Authors: Adam J Shephard, Raja Muhammad Saad Bashir, Hanya Mahmood, Mostafa Jahanifar, Fayyaz Minhas, Shan E Ahmed Raza, Kris D McCombe, Stephanie G Craig, Jacqueline James, Jill Brooks, Paul Nankivell, Hisham Mehanna, Syed Ali Khurram, Nasir M Rajpoot
Oral epithelial dysplasia (OED) is a premalignant histopathological diagnosis given to lesions of the oral cavity. Its grading suffers from significant inter-/intra- observer variability, and does not reliably predict malignancy progression, potentially leading to suboptimal treatment decisions. To address this, we developed a novel artificial intelligence algorithm that can assign an Oral Malignant Transformation (OMT) risk score, based on histological patterns in the in Haematoxylin and Eosin stained whole slide images, to quantify the risk of OED progression. The algorithm is based on the detection and segmentation of nuclei within (and around) the epithelium using an in-house segmentation model. We then employed a shallow neural network fed with interpretable morphological/spatial features, emulating histological markers. We conducted internal cross-validation on our development cohort (Sheffield; n = 193 cases) followed by independent validation on two external cohorts (Birmingham and Belfast; n = 92 cases). The proposed OMTscore yields an AUROC = 0.74 in predicting whether an OED progresses to malignancy or not. Survival analyses showed the prognostic value of our OMTscore for predicting malignancy transformation, when compared to the manually-assigned WHO and binary grades. Analysis of the correctly predicted cases elucidated the presence of peri-epithelial and epithelium-infiltrating lymphocytes in the most predictive patches of cases that transformed (p < 0.0001). This is the first study to propose a completely automated algorithm for predicting OED transformation based on interpretable nuclear features, whilst being validated on external datasets. The algorithm shows better-than-human-level performance for prediction of OED malignant transformation and offers a promising solution to the challenges of grading OED in routine clinical practice.
Authors: Mark S. Graham, Walter Hugo Lopez Pinaya, Paul Wright, Petru-Daniel Tudosiu, Yee H. Mah, James T. Teo, H. Rolf Jäger, David Werring, Parashkev Nachev, Sebastien Ourselin, M. Jorge Cardoso
Methods for out-of-distribution (OOD) detection that scale to 3D data are crucial components of any real-world clinical deep learning system. Classic denoising diffusion probabilistic models (DDPMs) have been recently proposed as a robust way to perform reconstruction-based OOD detection on 2D datasets, but do not trivially scale to 3D data. In this work, we propose to use Latent Diffusion Models (LDMs), which enable the scaling of DDPMs to high-resolution 3D medical data. We validate the proposed approach on near- and far-OOD datasets and compare it to a recently proposed, 3D-enabled approach using Latent Transformer Models (LTMs). Not only does the proposed LDM-based approach achieve statistically significant better performance, it also shows less sensitivity to the underlying latent representation, more favourable memory scaling, and produces better spatial anomaly maps. Code is available at https://github.com/marksgraham/ddpm-ood
Authors: Haleh Damirchi, Michael Greenspan, Ali Etemad
We propose a novel solution for predicting future trajectories of pedestrians. Our method uses a multimodal encoder-decoder transformer architecture, which takes as input both pedestrian locations and ego-vehicle speeds. Notably, our decoder predicts the entire future trajectory in a single-pass and does not perform one-step-ahead prediction, which makes the method effective for embedded edge deployment. We perform detailed experiments and evaluate our method on two popular datasets, PIE and JAAD. Quantitative results demonstrate the superiority of our proposed model over the current state-of-the-art, which consistently achieves the lowest error for 3 time horizons of 0.5, 1.0 and 1.5 seconds. Moreover, the proposed method is significantly faster than the state-of-the-art for the two datasets of PIE and JAAD. Lastly, ablation experiments demonstrate the impact of the key multimodal configuration of our method.
Authors: Debasmita Pal, Arun Ross
Plant phenology and phenotype prediction using remote sensing data is increasingly gaining the attention of the plant science community to improve agricultural productivity. In this work, we generate synthetic forestry images that satisfy certain phenotypic attributes, viz. canopy greenness. The greenness index of plants describes a particular vegetation type in a mixed forest. Our objective is to develop a Generative Adversarial Network (GAN) to synthesize forestry images conditioned on this continuous attribute, i.e., greenness of vegetation, over a specific region of interest. The training data is based on the automated digital camera imagery provided by the National Ecological Observatory Network (NEON) and processed by the PhenoCam Network. The synthetic images generated by our method are also used to predict another phenotypic attribute, viz., redness of plants. The Structural SIMilarity (SSIM) index is utilized to assess the quality of the synthetic images. The greenness and redness indices of the generated synthetic images are compared against that of the original images using Root Mean Squared Error (RMSE) in order to evaluate their accuracy and integrity. Moreover, the generalizability and scalability of our proposed GAN model is determined by effectively transforming it to generate synthetic images for other forest sites and vegetation types.
Authors: Matthias Freiberger, Peter Kun, Anders Sundnes Løvlie, Sebastian Risi
Models leveraging both visual and textual data such as Contrastive Language-Image Pre-training (CLIP), are increasingly gaining importance. In this work, we show that despite their versatility, such models are vulnerable to what we refer to as fooling master images. Fooling master images are capable of maximizing the confidence score of a CLIP model for a significant number of widely varying prompts, while being unrecognizable for humans. We demonstrate how fooling master images can be mined by searching the latent space of generative models by means of an evolution strategy or stochastic gradient descent. We investigate the properties of the mined fooling master images, and find that images trained on a small number of image captions potentially generalize to a much larger number of semantically related captions. Further, we evaluate two possible mitigation strategies and find that vulnerability to fooling master examples is closely related to a modality gap in contrastive pre-trained multi-modal networks. From the perspective of vulnerability to off-manifold attacks, we therefore argue for the mitigation of modality gaps in CLIP and related multi-modal approaches. Source code and mined CLIPMasterPrints are available at https://github.com/matfrei/CLIPMasterPrints.
Authors: Zhongliang Jiang, Chenyang Li, Xuesong Li, Nassir Navab
Autonomous ultrasound (US) imaging has gained increased interest recently, and it has been seen as a potential solution to overcome the limitations of free-hand US examinations, such as inter-operator variations. However, it is still challenging to accurately map planned paths from a generic atlas to individual patients, particularly for thoracic applications with high acoustic-impedance bone structures under the skin. To address this challenge, a graph-based non-rigid registration is proposed to enable transferring planned paths from the atlas to the current setup by explicitly considering subcutaneous bone surface features instead of the skin surface. To this end, the sternum and cartilage branches are segmented using a template matching to assist coarse alignment of US and CT point clouds. Afterward, a directed graph is generated based on the CT template. Then, the self-organizing map using geographical distance is successively performed twice to extract the optimal graph representations for CT and US point clouds, individually. To evaluate the proposed approach, five cartilage point clouds from distinct patients are employed. The results demonstrate that the proposed graph-based registration can effectively map trajectories from CT to the current setup for displaying US views through limited intercostal space. The non-rigid registration results in terms of Hausdorff distance (Mean$\pm$SD) is 9.48$\pm$0.27 mm and the path transferring error in terms of Euclidean distance is 2.21$\pm$1.11 mm.
Authors: Abdollah Ghazvanchahi, Pejman Jahbedar Maralani, Alan R. Moody, April Khademi
Deep learning (DL) methods for white matter lesion (WML) segmentation in MRI suffer a reduction in performance when applied on data from a scanner or centre that is out-of-distribution (OOD) from the training data. This is critical for translation and widescale adoption, since current models cannot be readily applied to data from new institutions. In this work, we evaluate several intensity standardization methods for MRI as a preprocessing step for WML segmentation in multi-centre Fluid-Attenuated Inversion Recovery (FLAIR) MRI. We evaluate a method specifically developed for FLAIR MRI called IAMLAB along with other popular normalization techniques such as White-strip, Nyul and Z-score. We proposed an Ensemble model that combines predictions from each of these models. A skip-connection UNet (SC UNet) was trained on the standardized images, as well as the original data and segmentation performance was evaluated over several dimensions. The training (in-distribution) data consists of a single study, of 60 volumes, and the test (OOD) data is 128 unseen volumes from three clinical cohorts. Results show IAMLAB and Ensemble provide higher WML segmentation performance compared to models from original data or other normalization methods. IAMLAB & Ensemble have the highest dice similarity coefficient (DSC) on the in-distribution data (0.78 & 0.80) and on clinical OOD data. DSC was significantly higher for IAMLAB compared to the original data (p<0.05) for all lesion categories (LL>25mL: 0.77 vs. 0.71; 10mL<= LL<25mL: 0.66 vs. 0.61; LL<10mL: 0.53 vs. 0.52). The IAMLAB and Ensemble normalization methods are mitigating MRI domain shift and are optimal for DL-based WML segmentation in unseen FLAIR data.
Authors: Zhongyu Jiang, Zhuoran Zhou, Lei Li, Wenhao Chai, Cheng-Yen Yang, Jenq-Neng Hwang
Learning-based methods have dominated the 3D human pose estimation (HPE) tasks with significantly better performance in most benchmarks than traditional optimization-based methods. Nonetheless, 3D HPE in the wild is still the biggest challenge of learning-based models, whether with 2D-3D lifting, image-to-3D, or diffusion-based methods, since the trained networks implicitly learn camera intrinsic parameters and domain-based 3D human pose distributions and estimate poses by statistical average. On the other hand, the optimization-based methods estimate results case-by-case, which can predict more diverse and sophisticated human poses in the wild. By combining the advantages of optimization-based and learning-based methods, we propose the Zero-shot Diffusion-based Optimization (ZeDO) pipeline for 3D HPE to solve the problem of cross-domain and in-the-wild 3D HPE. Our multi-hypothesis ZeDO achieves state-of-the-art (SOTA) performance on Human3.6M as minMPJPE $51.4$mm without training with any 2D-3D or image-3D pairs. Moreover, our single-hypothesis ZeDO achieves SOTA performance on 3DPW dataset with PA-MPJPE $42.6$mm on cross-dataset evaluation, which even outperforms learning-based methods trained on 3DPW.
Authors: Vaibhav Vavilala, Seemandhar Jain, Rahul Vasanth, Anand Bhattad, David Forsyth
We present Blocks2World, a novel method for 3D scene rendering and editing that leverages a two-step process: convex decomposition of images and conditioned synthesis. Our technique begins by extracting 3D parallelepipeds from various objects in a given scene using convex decomposition, thus obtaining a primitive representation of the scene. These primitives are then utilized to generate paired data through simple ray-traced depth maps. The next stage involves training a conditioned model that learns to generate images from the 2D-rendered convex primitives. This step establishes a direct mapping between the 3D model and its 2D representation, effectively learning the transition from a 3D model to an image. Once the model is fully trained, it offers remarkable control over the synthesis of novel and edited scenes. This is achieved by manipulating the primitives at test time, including translating or adding them, thereby enabling a highly customizable scene rendering process. Our method provides a fresh perspective on 3D scene rendering and editing, offering control and flexibility. It opens up new avenues for research and applications in the field, including authoring and data augmentation.
Authors: Zahid Hasan, Abu Zaher Md Faridee, Masud Ahmed, Sanjay Purushotham, Heesung Kwon, Hyungtae Lee, Nirmalya Roy
Novel Categories Discovery (NCD) tackles the open-world problem of classifying known and clustering novel categories based on the class semantics using partial class space annotated data. Unlike traditional pseudo-label and retraining, we investigate NCD from the novel data probability matrix perspective. We leverage the connection between NCD novel data sampling with provided novel class Multinoulli (categorical) distribution and hypothesize to implicitly achieve semantic-based novel data clustering by learning their class distribution. We propose novel constraints on first-order (mean) and second-order (covariance) statistics of probability matrix features while applying instance-wise information constraints. In particular, we align the neuron distribution (activation patterns) under a large batch of Monte-Carlo novel data sampling by matching their empirical features mean and covariance with the provided Multinoulli-distribution. Simultaneously, we minimize entropy and enforce prediction consistency for each instance. Our simple approach successfully realizes semantic-based novel data clustering provided the semantic similarity between label-unlabeled classes. We demonstrate the discriminative capacity of our approaches in image and video modalities. Moreover, we perform extensive ablation studies regarding data, networks, and our framework components to provide better insights. Our approach maintains ~94%, ~93%, and ~85%, classification accuracy in labeled data while achieving ~90%, ~84%, and ~72% clustering accuracy for novel categories for Cifar10, UCF101, and MPSC-ARL datasets that matches state-of-the-art approaches without any external clustering.
Authors: Aditya Sanghi, Pradeep Kumar Jayaraman, Arianna Rampini, Joseph Lambourne, Hooman Shayani, Evan Atherton, Saeid Asgari Taghanaki
Significant progress has recently been made in creative applications of large pre-trained models for downstream tasks in 3D vision, such as text-to-shape generation. This motivates our investigation of how these pre-trained models can be used effectively to generate 3D shapes from sketches, which has largely remained an open challenge due to the limited sketch-shape paired datasets and the varying level of abstraction in the sketches. We discover that conditioning a 3D generative model on the features (obtained from a frozen large pre-trained vision model) of synthetic renderings during training enables us to effectively generate 3D shapes from sketches at inference time. This suggests that the large pre-trained vision model features carry semantic signals that are resilient to domain shifts, i.e., allowing us to use only RGB renderings, but generalizing to sketches at inference time. We conduct a comprehensive set of experiments investigating different design factors and demonstrate the effectiveness of our straightforward approach for generation of multiple 3D shapes per each input sketch regardless of their level of abstraction without requiring any paired datasets during training.
Authors: Dhiraj Neupane, Lakpa Dorje Tamang, Ngoc Dung Huynh, Mohamed Reda Bouadjenek, Sunil Aryal
We implemented a simple method for early detection in this research. The implemented methods are plotting the given mat files and analyzing scalogram images generated by performing Continuous Wavelet Transform (CWT) on the samples. Also, finding the mean, standard deviation (STD), and peak-to-peak (P2P) values from each signal also helped detect faulty signs. We have implemented the autoregressive integrated moving average (ARIMA) method to track the progression.
Authors: Amanda Dy, Ngoc-Nhu Jennifer Nguyen, Seyed Hossein Mirjahanmardi, Melanie Dawe, Anthony Fyles, Wei Shi, Fei-Fei Liu, Dimitrios Androutsos, Susan Done, April Khademi
Deep learning systems have been proposed to improve the objectivity and efficiency of Ki- 67 PI scoring. The challenge is that while very accurate, deep learning techniques suffer from reduced performance when applied to out-of-domain data. This is a critical challenge for clinical translation, as models are typically trained using data available to the vendor, which is not from the target domain. To address this challenge, this study proposes a domain adaptation pipeline that employs an unsupervised framework to generate silver standard (pseudo) labels in the target domain, which is used to augment the gold standard (GS) source domain data. Five training regimes were tested on two validated Ki-67 scoring architectures (UV-Net and piNET), (1) SS Only: trained on target silver standard (SS) labels, (2) GS Only: trained on source GS labels, (3) Mixed: trained on target SS and source GS labels, (4) GS+SS: trained on source GS labels and fine-tuned on target SS labels, and our proposed method (5) SS+GS: trained on source SS labels and fine-tuned on source GS labels. The SS+GS method yielded significantly (p < 0.05) higher PI accuracy (95.9%) and more consistent results compared to the GS Only model on target data. Analysis of t-SNE plots showed features learned by the SS+GS models are more aligned for source and target data, resulting in improved generalization. The proposed pipeline provides an efficient method for learning the target distribution without manual annotations, which are time-consuming and costly to generate for medical images. This framework can be applied to any target site as a per-laboratory calibration method, for widescale deployment.
Authors: Robin Netzorg, Jiaxun Li, Bin Yu
In recent years, work has gone into developing deep interpretable methods for image classification that clearly attributes a model's output to specific features of the data. One such of these methods is the prototypical part network (ProtoPNet), which attempts to classify images based on meaningful parts of the input. While this method results in interpretable classifications, this method often learns to classify from spurious or inconsistent parts of the image. Hoping to remedy this, we take inspiration from the recent developments in Reinforcement Learning with Human Feedback (RLHF) to fine-tune these prototypes. By collecting human annotations of prototypes quality via a 1-5 scale on the CUB-200-2011 dataset, we construct a reward model that learns to identify non-spurious prototypes. In place of a full RL update, we propose the reweighted, reselected, and retrained prototypical part network (R3-ProtoPNet), which adds an additional three steps to the ProtoPNet training loop. The first two steps are reward-based reweighting and reselection, which align prototypes with human feedback. The final step is retraining to realign the model's features with the updated prototypes. We find that R3-ProtoPNet improves the overall consistency and meaningfulness of the prototypes, but lower the test predictive accuracy when used independently. When multiple R3-ProtoPNets are incorporated into an ensemble, we find an increase in test predictive performance while maintaining interpretability.
Authors: Tianlei Zhu, Junqi Chen, Renzhe Zhu, Gaurav Gupta
StyleGAN can use style to affect facial posture and identity features, and noise to affect hair, wrinkles, skin color and other details. Among these, the outcomes of the picture processing will vary slightly between different versions of styleGAN. As a result, the comparison of performance differences between styleGAN2 and the two modified versions of styleGAN3 will be the main focus of this study. We used the FFHQ dataset as the dataset and FID, EQ-T, and EQ-R were used to be the assessment of the model. In the end, we discovered that Stylegan3 version is a better generative network to improve the equivariance. Our findings have a positive impact on the creation of animation and videos.
Authors: Huafeng Li, Le Xu, Yafei Zhang, Dapeng Tao, Zhengtao Yu
In visible-infrared video person re-identification (re-ID), extracting features not affected by complex scenes (such as modality, camera views, pedestrian pose, background, etc.) changes, and mining and utilizing motion information are the keys to solving cross-modal pedestrian identity matching. To this end, the paper proposes a new visible-infrared video person re-ID method from a novel perspective, i.e., adversarial self-attack defense and spatial-temporal relation mining. In this work, the changes of views, posture, background and modal discrepancy are considered as the main factors that cause the perturbations of person identity features. Such interference information contained in the training samples is used as an adversarial perturbation. It performs adversarial attacks on the re-ID model during the training to make the model more robust to these unfavorable factors. The attack from the adversarial perturbation is introduced by activating the interference information contained in the input samples without generating adversarial samples, and it can be thus called adversarial self-attack. This design allows adversarial attack and defense to be integrated into one framework. This paper further proposes a spatial-temporal information-guided feature representation network to use the information in video sequences. The network cannot only extract the information contained in the video-frame sequences but also use the relation of the local information in space to guide the network to extract more robust features. The proposed method exhibits compelling performance on large-scale cross-modality video datasets. The source code of the proposed method will be released at https://github.com/lhf12278/xxx.
Authors: Congqi Cao, Ze Sun, Qinyi Lv, Lingtong Min, Yanning Zhang
Egocentric action anticipation is a challenging task that aims to make advanced predictions of future actions from current and historical observations in the first-person view. Most existing methods focus on improving the model architecture and loss function based on the visual input and recurrent neural network to boost the anticipation performance. However, these methods, which merely consider visual information and rely on a single network architecture, gradually reach a performance plateau. In order to fully understand what has been observed and capture the dependencies between current observations and future actions well enough, we propose a novel visual-semantic fusion enhanced and Transformer GRU-based action anticipation framework in this paper. Firstly, high-level semantic information is introduced to improve the performance of action anticipation for the first time. We propose to use the semantic features generated based on the class labels or directly from the visual observations to augment the original visual features. Secondly, an effective visual-semantic fusion module is proposed to make up for the semantic gap and fully utilize the complementarity of different modalities. Thirdly, to take advantage of both the parallel and autoregressive models, we design a Transformer based encoder for long-term sequential modeling and a GRU-based decoder for flexible iteration decoding. Extensive experiments on two large-scale first-person view datasets, i.e., EPIC-Kitchens and EGTEA Gaze+, validate the effectiveness of our proposed method, which achieves new state-of-the-art performance, outperforming previous approaches by a large margin.
Authors: Dongyue Sun, Shiyao Jiang, Lin Qi
Existing edge-aware camouflaged object detection (COD) methods normally output the edge prediction in the early stage. However, edges are important and fundamental factors in the following segmentation task. Due to the high visual similarity between camouflaged targets and the surroundings, edge prior predicted in early stage usually introduces erroneous foreground-background and contaminates features for segmentation. To tackle this problem, we propose a novel Edge-aware Mirror Network (EAMNet), which models edge detection and camouflaged object segmentation as a cross refinement process. More specifically, EAMNet has a two-branch architecture, where a segmentation-induced edge aggregation module and an edge-induced integrity aggregation module are designed to cross-guide the segmentation branch and edge detection branch. A guided-residual channel attention module which leverages the residual connection and gated convolution finally better extracts structural details from low-level features. Quantitative and qualitative experiment results show that EAMNet outperforms existing cutting-edge baselines on three widely used COD datasets. Codes are available at https://github.com/sdy1999/EAMNet.
Authors: Yi Zhong, Mengqiu Xu, Kongming Liang, Kaixin Chen, Ming Wu
Segmentation of the infected areas of the lung is essential for quantifying the severity of lung disease like pulmonary infections. Existing medical image segmentation methods are almost uni-modal methods based on image. However, these image-only methods tend to produce inaccurate results unless trained with large amounts of annotated data. To overcome this challenge, we propose a language-driven segmentation method that uses text prompt to improve to the segmentation result. Experiments on the QaTa-COV19 dataset indicate that our method improves the Dice score by 6.09% at least compared to the uni-modal methods. Besides, our extended study reveals the flexibility of multi-modal methods in terms of the information granularity of text and demonstrates that multi-modal methods have a significant advantage over image-only methods in terms of the size of training data required.
Authors: Yuxuan Song, Xinyue Li, Lin Qi
The task of Camouflaged Object Detection (COD) aims to accurately segment camouflaged objects that integrated into the environment, which is more challenging than ordinary detection as the texture between the target and background is visually indistinguishable. In this paper, we proposed a novel Feature Grafting and Distractor Aware network (FDNet) to handle the COD task. Specifically, we use CNN and Transformer to encode multi-scale images in parallel. In order to better explore the advantages of the two encoders, we design a cross-attention-based Feature Grafting Module to graft features extracted from Transformer branch into CNN branch, after which the features are aggregated in the Feature Fusion Module. A Distractor Aware Module is designed to explicitly model the two possible distractors in the COD task to refine the coarse camouflage map. We also proposed the largest artificial camouflaged object dataset which contains 2000 images with annotations, named ACOD2K. We conducted extensive experiments on four widely used benchmark datasets and the ACOD2K dataset. The results show that our method significantly outperforms other state-of-the-art methods. The code and the ACOD2K will be available at https://github.com/syxvision/FDNet.
Authors: George Tom, Minesh Mathew, Sergi Garcia, Dimosthenis Karatzas, C.V. Jawahar
Text and signs around roads provide crucial information for drivers, vital for safe navigation and situational awareness. Scene text recognition in motion is a challenging problem, while textual cues typically appear for a short time span, and early detection at a distance is necessary. Systems that exploit such information to assist the driver should not only extract and incorporate visual and textual cues from the video stream but also reason over time. To address this issue, we introduce RoadTextVQA, a new dataset for the task of video question answering (VideoQA) in the context of driver assistance. RoadTextVQA consists of $3,222$ driving videos collected from multiple countries, annotated with $10,500$ questions, all based on text or road signs present in the driving videos. We assess the performance of state-of-the-art video question answering models on our RoadTextVQA dataset, highlighting the significant potential for improvement in this domain and the usefulness of the dataset in advancing research on in-vehicle support systems and text-aware multimodal question answering. The dataset is available at this http URL
Authors: Ahmad Sajedi, Samir Khaki, Konstantinos N. Plataniotis, Mahdi S. Hosseini
Multilabel representation learning is recognized as a challenging problem that can be associated with either label dependencies between object categories or data-related issues such as the inherent imbalance of positive/negative samples. Recent advances address these challenges from model- and data-centric viewpoints. In model-centric, the label correlation is obtained by an external model designs (e.g., graph CNN) to incorporate an inductive bias for training. However, they fail to design an end-to-end training framework, leading to high computational complexity. On the contrary, in data-centric, the realistic nature of the dataset is considered for improving the classification while ignoring the label dependencies. In this paper, we propose a new end-to-end training framework -- dubbed KMCL (Kernel-based Mutlilabel Contrastive Learning) -- to address the shortcomings of both model- and data-centric designs. The KMCL first transforms the embedded features into a mixture of exponential kernels in Gaussian RKHS. It is then followed by encoding an objective loss that is comprised of (a) reconstruction loss to reconstruct kernel representation, (b) asymmetric classification loss to address the inherent imbalance problem, and (c) contrastive loss to capture label correlation. The KMCL models the uncertainty of the feature encoder while maintaining a low computational footprint. Extensive experiments are conducted on image classification tasks to showcase the consistent improvements of KMCL over the SOTA methods. PyTorch implementation is provided in \url{https://github.com/mahdihosseini/KMCL}.
Authors: Sahil Gangurde
In city, information about urban objects such as water supply, railway lines, power lines, buildings, roads, etc., is necessary for city planning. In particular, information about the spread of these objects, locations and capacity is needed for the policymakers to make impactful decisions. This thesis aims to segment the building and roads from the aerial image captured by the satellites and UAVs. Many different architectures have been proposed for the semantic segmentation task and UNet being one of them. In this thesis, we propose a novel architecture based on Google's newly proposed EfficientNetV2 as an encoder for feature extraction with UNet decoder for constructing the segmentation map. Using this approach we achieved a benchmark score for the Massachusetts Building and Road dataset with an mIOU of 0.8365 and 0.9153 respectively.
Authors: Yuqian Chen, Leo R. Zekelman, Chaoyi Zhang, Tengfei Xue, Yang Song, Nikos Makris, Yogesh Rathi, Alexandra J. Golby, Weidong Cai, Fan Zhang, Lauren J. O'Donnell
We propose a geometric deep-learning-based framework, TractGeoNet, for performing regression using diffusion magnetic resonance imaging (dMRI) tractography and associated pointwise tissue microstructure measurements. By employing a point cloud representation, TractGeoNet can directly utilize pointwise tissue microstructure and positional information from all points within a fiber tract. To improve regression performance, we propose a novel loss function, the Paired-Siamese Regression loss, which encourages the model to focus on accurately predicting the relative differences between regression label scores rather than just their absolute values. In addition, we propose a Critical Region Localization algorithm to identify highly predictive anatomical regions within the white matter fiber tracts for the regression task. We evaluate the effectiveness of the proposed method by predicting individual performance on two neuropsychological assessments of language using a dataset of 20 association white matter fiber tracts from 806 subjects from the Human Connectome Project. The results demonstrate superior prediction performance of TractGeoNet compared to several popular regression models. Of the twenty tracts studied, we find that the left arcuate fasciculus tract is the most highly predictive of the two studied language performance assessments. The localized critical regions are widespread and distributed across both hemispheres and all cerebral lobes, including areas of the brain considered important for language function such as superior and anterior temporal regions, pars opercularis, and precentral gyrus. Overall, TractGeoNet demonstrates the potential of geometric deep learning to enhance the study of the brain's white matter fiber tracts and to relate their structure to human traits such as language performance.
Authors: Ganglai Wang, Peng Zhang, Junwen Xiong, Feihan Yang, Wei Huang, Yufei Zha
DeepFake based digital facial forgery is threatening public media security, especially when lip manipulation has been used in talking face generation, and the difficulty of fake video detection is further improved. By only changing lip shape to match the given speech, the facial features of identity are hard to be discriminated in such fake talking face videos. Together with the lack of attention on audio stream as the prior knowledge, the detection failure of fake talking face videos also becomes inevitable. It's found that the optical flow of the fake talking face video is disordered especially in the lip region while the optical flow of the real video changes regularly, which means the motion feature from optical flow is useful to capture manipulation cues. In this study, a fake talking face detection network (FTFDNet) is proposed by incorporating visual, audio and motion features using an efficient cross-modal fusion (CMF) module. Furthermore, a novel audio-visual attention mechanism (AVAM) is proposed to discover more informative features, which can be seamlessly integrated into any audio-visual CNN architecture by modularization. With the additional AVAM, the proposed FTFDNet is able to achieve a better detection performance than other state-of-the-art DeepFake video detection methods not only on the established fake talking face detection dataset (FTFDD) but also on the DeepFake video detection datasets (DFDC and DF-TIMIT).
Authors: Tong Li, Hansen Feng, Lizhi Wang, Zhiwei Xiong, Hua Huang
Image denoising is a fundamental problem in computational photography, where achieving high-quality perceptual performance with low distortion is highly demanding. Current methods either struggle with perceptual performance or suffer from significant distortion. Recently, the emerging diffusion model achieves state-of-the-art performance in various tasks, and its denoising mechanism demonstrates great potential for image denoising. However, stimulating diffusion models for image denoising is not straightforward and requires solving several critical problems. On the one hand, the input inconsistency hinders the connection of diffusion models and image denoising. On the other hand, the content inconsistency between the generated image and the desired denoised image introduces additional distortion. To tackle these problems, we present a novel strategy called Diffusion Model for Image Denoising (DMID) by understanding and rethinking the diffusion model from a denoising perspective. Our DMID strategy includes an adaptive embedding method that embeds the noisy image into a pre-trained diffusion model, and an adaptive ensembling method that reduces distortion in the denoised image. Our DMID strategy achieves state-of-the-art performance on all distortion-based and perceptual metrics, for both Gaussian and real-world image denoising.
Authors: Liqi Xue, Tianyi Xu, Yongbao Song, Yan Liu, Lei Zhang, Xiantong Zhen, Jun Xu
The display devices like HDR10 televisions are increasingly prevalent in our daily life for visualizing high dynamic range (HDR) images. But the majority of media images on the internet remain in 8-bit standard dynamic range (SDR) format. Therefore, converting SDR images to HDR ones by inverse tone mapping (ITM) is crucial to unlock the full potential of abundant media images. However, existing ITM methods are usually developed with complex network architectures requiring huge computational costs. In this paper, we propose a lightweight Improved Residual Network (IRNet) by enhancing the power of popular residual block for efficient ITM. Specifically, we propose a new Improved Residual Block (IRB) to extract and fuse multi-layer features for fine-grained HDR image reconstruction. Experiments on three benchmark datasets demonstrate that our IRNet achieves state-of-the-art performance on both the ITM and joint SR-ITM tasks. The code, models and data will be publicly available at https://github.com/ThisisVikki/ITM-baseline.
Authors: Rao Fu, Cheng Wen, Qian Li, Xiao Xiao, Pierre Alliez
This paper proposes BPNet, a novel end-to-end deep learning framework to learn B\'ezier primitive segmentation on 3D point clouds. The existing works treat different primitive types separately, thus limiting them to finite shape categories. To address this issue, we seek a generalized primitive segmentation on point clouds. Taking inspiration from B\'ezier decomposition on NURBS models, we transfer it to guide point cloud segmentation casting off primitive types. A joint optimization framework is proposed to learn B\'ezier primitive segmentation and geometric fitting simultaneously on a cascaded architecture. Specifically, we introduce a soft voting regularizer to improve primitive segmentation and propose an auto-weight embedding module to cluster point features, making the network more robust and generic. We also introduce a reconstruction module where we successfully process multiple CAD models with different primitives simultaneously. We conducted extensive experiments on the synthetic ABC dataset and real-scan datasets to validate and compare our approach with different baseline methods. Experiments show superior performance over previous work in terms of segmentation, with a substantially faster inference speed.
Authors: Amirhossein Askari-Farsangi, Ali Sharifi-Zarchi, Mohammad Hossein Rohban
Acute Lymphoblastic Leukemia (ALL) is one of the most common types of childhood blood cancer. The quick start of the treatment process is critical to saving the patient's life, and for this reason, early diagnosis of this disease is essential. Examining the blood smear images of these patients is one of the methods used by expert doctors to diagnose this disease. Deep learning-based methods have numerous applications in medical fields, as they have significantly advanced in recent years. ALL diagnosis is not an exception in this field, and several machine learning-based methods for this problem have been proposed. In previous methods, high diagnostic accuracy was reported, but our work showed that this alone is not sufficient, as it can lead to models taking shortcuts and not making meaningful decisions. This issue arises due to the small size of medical training datasets. To address this, we constrained our model to follow a pipeline inspired by experts' work. We also demonstrated that, since a judgement based on only one image is insufficient, redefining the problem as a multiple-instance learning problem is necessary for achieving a practical result. Our model is the first to provide a solution to this problem in a multiple-instance learning setup. We introduced a novel pipeline for diagnosing ALL that approximates the process used by hematologists, is sensitive to disease biomarkers, and achieves an accuracy of 96.15%, an F1-score of 94.24%, a sensitivity of 97.56%, and a specificity of 90.91% on ALL IDB 1. Our method was further evaluated on an out-of-distribution dataset, which posed a challenging test and had acceptable performance. Notably, our model was trained on a relatively small dataset, highlighting the potential for our approach to be applied to other medical datasets with limited data availability.
Authors: Stephen Casper, Zifan Guo, Shreya Mogulothu, Zachary Marinov, Chinmay Deshpande, Rui-Jie Yew, Zheng Dai, Dylan Hadfield-Menell
Modern diffusion models have set the state-of-the-art in AI image generation. Their success is due, in part, to training on Internet-scale data which often includes copyrighted work. This prompts questions about the extent to which these models learn from, imitate, or copy the work of human artists. This work suggests that tying copyright liability to the capabilities of the model may be useful given the evolving ecosystem of generative models. Specifically, much of the legal analysis of copyright and generative systems focuses on the use of protected data for training. As a result, the connections between data, training, and the system are often obscured. In our approach, we consider simple image classification techniques to measure a model's ability to imitate specific artists. Specifically, we use Contrastive Language-Image Pretrained (CLIP) encoders to classify images in a zero-shot fashion. Our process first prompts a model to imitate a specific artist. Then, we test whether CLIP can be used to reclassify the artist (or the artist's work) from the imitation. If these tests match the imitation back to the original artist, this suggests the model can imitate that artist's expression. Our approach is simple and quantitative. Furthermore, it uses standard techniques and does not require additional training. We demonstrate our approach with an audit of Stable Diffusion's capacity to imitate 70 professional digital artists with copyrighted work online. When Stable Diffusion is prompted to imitate an artist from this set, we find that the artist can be identified from the imitation with an average accuracy of 81.0%. Finally, we also show that a sample of the artist's work can be matched to these imitation images with a high degree of statistical reliability. Overall, these results suggest that Stable Diffusion is broadly successful at imitating individual human artists.
Authors: Tong Steven Sun, Yuyang Gao, Shubham Khaladkar, Sijia Liu, Liang Zhao, Young-Ho Kim, Sungsoo Ray Hong
The local explanation provides heatmaps on images to explain how Convolutional Neural Networks (CNNs) derive their output. Due to its visual straightforwardness, the method has been one of the most popular explainable AI (XAI) methods for diagnosing CNNs. Through our formative study (S1), however, we captured ML engineers' ambivalent perspective about the local explanation as a valuable and indispensable envision in building CNNs versus the process that exhausts them due to the heuristic nature of detecting vulnerability. Moreover, steering the CNNs based on the vulnerability learned from the diagnosis seemed highly challenging. To mitigate the gap, we designed DeepFuse, the first interactive design that realizes the direct feedback loop between a user and CNNs in diagnosing and revising CNN's vulnerability using local explanations. DeepFuse helps CNN engineers to systemically search "unreasonable" local explanations and annotate the new boundaries for those identified as unreasonable in a labor-efficient manner. Next, it steers the model based on the given annotation such that the model doesn't introduce similar mistakes. We conducted a two-day study (S2) with 12 experienced CNN engineers. Using DeepFuse, participants made a more accurate and "reasonable" model than the current state-of-the-art. Also, participants found the way DeepFuse guides case-based reasoning can practically improve their current practice. We provide implications for design that explain how future HCI-driven design can move our practice forward to make XAI-driven insights more actionable.
Authors: Qin Zhang, Linghan Xu, Qingming Tang, Jun Fang, Ying Nian Wu, Joe Tighe, Yifan Xing
The ability to use the same distance threshold across different test classes / distributions is highly desired for a frictionless deployment of commercial image retrieval systems. However, state-of-the-art deep metric learning losses often result in highly varied intra-class and inter-class embedding structures, making threshold calibration a non-trivial process in practice. In this paper, we propose a novel metric named Operating-Point-Incosistency-Score (OPIS) that measures the variance in the operating characteristics across different classes in a target calibration range, and demonstrate that high accuracy of a metric learning embedding model does not guarantee calibration consistency for both seen and unseen classes. We find that, in the high-accuracy regime, there exists a Pareto frontier where accuracy improvement comes at the cost of calibration consistency. To address this, we develop a novel regularization, named Calibration-Aware Margin (CAM) loss, to encourage uniformity in the representation structures across classes during training. Extensive experiments demonstrate CAM's effectiveness in improving calibration-consistency while retaining or even enhancing accuracy, outperforming state-of-the-art deep metric learning methods.
Authors: Sen Lu, Abhronil Sengupta
Spike-Timing-Dependent Plasticity (STDP) is an unsupervised learning mechanism for Spiking Neural Networks (SNNs) that has received significant attention from the neuromorphic hardware community. However, scaling such local learning techniques to deeper networks and large-scale tasks has remained elusive. In this work, we investigate a Deep-STDP framework where a convolutional network is trained in tandem with pseudo-labels generated by the STDP clustering process on the network outputs. We achieve $24.56\%$ higher accuracy and $3.5\times$ faster convergence speed at iso-accuracy on a 10-class subset of the Tiny ImageNet dataset in contrast to a $k$-means clustering approach.
Authors: Mingzhen Shao
Previous studies have shown the vulnerability of vision transformers to adversarial patches, but these studies all rely on a critical assumption: the attack patches must be perfectly aligned with the patches used for linear projection in vision transformers. Due to this stringent requirement, deploying adversarial patches for vision transformers in the physical world becomes impractical, unlike their effectiveness on CNNs. This paper proposes a novel method for generating an adversarial patch (G-Patch) that overcomes the alignment constraint, allowing the patch to launch a targeted attack at any position within the field of view. Specifically, instead of directly optimizing the patch using gradients, we employ a GAN-like structure to generate the adversarial patch. Our experiments show the effectiveness of the adversarial patch in achieving universal attacks on vision transformers, both in digital and physical-world scenarios. Additionally, further analysis reveals that the generated adversarial patch exhibits robustness to brightness restriction, color transfer, and random noise. Real-world attack experiments validate the effectiveness of the G-Patch to launch robust attacks even under some very challenging conditions.
Authors: Paul Kuo-Ming Huang, Si-An Chen, Hsuan-Tien Lin
Score-based Generative Models (SGMs) are a popular family of deep generative models that achieves leading image generation quality. Earlier studies have extended SGMs to tackle class-conditional generation by coupling an unconditional SGM with the guidance of a trained classifier. Nevertheless, such classifier-guided SGMs do not always achieve accurate conditional generation, especially when trained with fewer labeled data. We argue that the issue is rooted in unreliable gradients of the classifier and the inability to fully utilize unlabeled data during training. We then propose to improve classifier-guided SGMs by letting the classifier calibrate itself. Our key idea is to use principles from energy-based models to convert the classifier as another view of the unconditional SGM. Then, existing loss for the unconditional SGM can be adopted to calibrate the classifier using both labeled and unlabeled data. Empirical results validate that the proposed approach significantly improves the conditional generation quality across different percentages of labeled data. The improved performance makes the proposed approach consistently superior to other conditional SGMs when using fewer labeled data. The results confirm the potential of the proposed approach for generative modeling with limited labeled data.
Authors: Bo Zhao, Boya Wu, Tiejun Huang
Thanks to the emerging of foundation models, the large language and vision models are integrated to acquire the multimodal ability of visual captioning, dialogue, question answering, etc. Although existing multimodal models present impressive performance of visual understanding and reasoning, their limits are still largely under-explored due to the scarcity of high-quality instruction tuning data. To push the limits of multimodal capability, we Sale up Visual Instruction Tuning (SVIT) by constructing a dataset of 3.2 million visual instruction tuning data including 1.6M conversation question-answer (QA) pairs and 1.6M complex reasoning QA pairs and 106K detailed image descriptions. Besides the volume, the proposed dataset is also featured by the high quality and rich diversity, which is generated by prompting GPT-4 with the abundant manual annotations of images. We empirically verify that training multimodal models on SVIT can significantly improve the multimodal performance in terms of visual perception, reasoning and planing.
Authors: Jun Cen, Shiwei Zhang, Yixuan Pei, Kun Li, Hang Zheng, Maochun Luo, Yingya Zhang, Qifeng Chen
2D RGB images and 3D LIDAR point clouds provide complementary knowledge for the perception system of autonomous vehicles. Several 2D and 3D fusion methods have been explored for the LIDAR semantic segmentation task, but they suffer from different problems. 2D-to-3D fusion methods require strictly paired data during inference, which may not be available in real-world scenarios, while 3D-to-2D fusion methods cannot explicitly make full use of the 2D information. Therefore, we propose a Bidirectional Fusion Network with Cross-Modality Knowledge Distillation (CMDFusion) in this work. Our method has two contributions. First, our bidirectional fusion scheme explicitly and implicitly enhances the 3D feature via 2D-to-3D fusion and 3D-to-2D fusion, respectively, which surpasses either one of the single fusion schemes. Second, we distillate the 2D knowledge from a 2D network (Camera branch) to a 3D network (2D knowledge branch) so that the 3D network can generate 2D information even for those points not in the FOV (field of view) of the camera. In this way, RGB images are not required during inference anymore since the 2D knowledge branch provides 2D information according to the 3D LIDAR input. We show that our CMDFusion achieves the best performance among all fusion-based methods on SemanticKITTI and nuScenes datasets. The code will be released at https://github.com/Jun-CEN/CMDFusion.
Authors: Tao Wu, Tie Luo, Donald C. Wunsch
Adversarial examples (AE) with good transferability enable practical black-box attacks on diverse target models, where insider knowledge about the target models is not required. Previous methods often generate AE with no or very limited transferability; that is, they easily overfit to the particular architecture and feature representation of the source, white-box model and the generated AE barely work for target, black-box models. In this paper, we propose a novel approach to enhance AE transferability using Gradient Norm Penalty (GNP). It drives the loss function optimization procedure to converge to a flat region of local optima in the loss landscape. By attacking 11 state-of-the-art (SOTA) deep learning models and 6 advanced defense methods, we empirically show that GNP is very effective in generating AE with high transferability. We also demonstrate that it is very flexible in that it can be easily integrated with other gradient based methods for stronger transfer-based attacks.
Authors: Nati Ofir
This paper addresses the problem of visible (RGB) to Near-Infrared (NIR) image fusion. Multispectral imaging is an important task relevant to image processing and computer vision, even more, since the development of the RGBT sensor. While the visible image sees color and suffers from noise, haze, and clouds, the NIR channel captures a clearer picture and it is significantly required by applications such as dehazing or object detection. The proposed approach fuses these two aligned channels by training a Convolutional-Neural-Network (CNN) by a Self-Supervised-Learning (SSL) on a single example. For each such pair, RGB and IR, the network is trained for seconds to deduce the final fusion. The SSL is based on Sturcture-of-Similarity (SSIM) loss combined with Edge-Preservation (EP) loss. The labels for the SSL are the input channels themselves. This fusion preserves the relevant detail of each spectral channel while not based on a heavy training process. In the experiments section, the proposed approach achieves better qualitative and quantitative multispectral fusion results with respect to other recent methods, that are not based on large dataset training.
Authors: Zhiling Guo, Xiaodan Shi, Haoran Zhang, Dou Huang, Xiaoya Song, Jinyue Yan, Ryosuke Shibasaki
The development of remote sensing and deep learning techniques has enabled building semantic segmentation with high accuracy and efficiency. Despite their success in different tasks, the discussions on the impact of spatial resolution on deep learning based building semantic segmentation are quite inadequate, which makes choosing a higher cost-effective data source a big challenge. To address the issue mentioned above, in this study, we create remote sensing images among three study areas into multiple spatial resolutions by super-resolution and down-sampling. After that, two representative deep learning architectures: UNet and FPN, are selected for model training and testing. The experimental results obtained from three cities with two deep learning models indicate that the spatial resolution greatly influences building segmentation results, and with a better cost-effectiveness around 0.3m, which we believe will be an important insight for data selection and preparation.
Authors: Zhijian Liu, Nian Cai, Wensheng Ouyang, Chengbin Zhang, Nili Tian, Han Wang
Automatic hardhat wearing detection can strengthen the safety management in construction sites, which is still challenging due to complicated video surveillance scenes. To deal with the poor generalization of previous deep learning based methods, a novel anchor-free deep learning framework called CA-CentripetalNet is proposed for hardhat wearing detection. Two novel schemes are proposed to improve the feature extraction and utilization ability of CA-CentripetalNet, which are vertical-horizontal corner pooling and bounding constrained center attention. The former is designed to realize the comprehensive utilization of marginal features and internal features. The latter is designed to enforce the backbone to pay attention to internal features, which is only used during the training rather than during the detection. Experimental results indicate that the CA-CentripetalNet achieves better performance with the 86.63% mAP (mean Average Precision) with less memory consumption at a reasonable speed than the existing deep learning based methods, especially in case of small-scale hardhats and non-worn-hardhats.
Authors: Jiayu Yang, Enze Xie, Jose M. Alvarez, Miaomiao Liu
Recent vision-only perception models for autonomous driving achieved promising results by encoding multi-view image features into Bird's-Eye-View (BEV) space. A critical step and the main bottleneck of these methods is transforming image features into the BEV coordinate frame. This paper focuses on leveraging geometry information, such as depth, to model such feature transformation. Existing works rely on non-parametric depth distribution modeling leading to significant memory consumption, or ignore the geometry information to address this problem. In contrast, we propose to use parametric depth distribution modeling for feature transformation. We first lift the 2D image features to the 3D space defined for the ego vehicle via a predicted parametric depth distribution for each pixel in each view. Then, we aggregate the 3D feature volume based on the 3D space occupancy derived from depth to the BEV frame. Finally, we use the transformed features for downstream tasks such as object detection and semantic segmentation. Existing semantic segmentation methods do also suffer from an hallucination problem as they do not take visibility information into account. This hallucination can be particularly problematic for subsequent modules such as control and planning. To mitigate the issue, our method provides depth uncertainty and reliable visibility-aware estimations. We further leverage our parametric depth modeling to present a novel visibility-aware evaluation metric that, when taken into account, can mitigate the hallucination problem. Extensive experiments on object detection and semantic segmentation on the nuScenes datasets demonstrate that our method outperforms existing methods on both tasks.
Authors: Kazuya Nishimura, Ami Katanaya, Shinichiro Chuma, Ryoma Bise
Detection of mitosis events plays an important role in biomedical research. Deep-learning-based mitosis detection methods have achieved outstanding performance with a certain amount of labeled data. However, these methods require annotations for each imaging condition. Collecting labeled data involves time-consuming human labor. In this paper, we propose a mitosis detection method that can be trained with partially annotated sequences. The base idea is to generate a fully labeled dataset from the partial labels and train a mitosis detection model with the generated dataset. First, we generate an image pair not containing mitosis events by frame-order flipping. Then, we paste mitosis events to the image pair by alpha-blending pasting and generate a fully labeled dataset. We demonstrate the performance of our method on four datasets, and we confirm that our method outperforms other comparisons which use partially labeled sequences.
Authors: Zihao Jiang, Yunkai Dang, Dong Pang, Huishuai Zhang, Weiran Huang
Few-shot learning aims to train models that can be generalized to novel classes with only a few samples. Recently, a line of works are proposed to enhance few-shot learning with accessible semantic information from class names. However, these works focus on improving existing modules such as visual prototypes and feature extractors of the standard few-shot learning framework. This limits the full potential use of semantic information. In this paper, we propose a novel few-shot learning framework that uses pre-trained language models based on contrastive learning. To address the challenge of alignment between visual features and textual embeddings obtained from text-based pre-trained language model, we carefully design the textual branch of our framework and introduce a metric module to generalize the cosine similarity. For better transferability, we let the metric module adapt to different few-shot tasks and adopt MAML to train the model via bi-level optimization. Moreover, we conduct extensive experiments on multiple benchmarks to demonstrate the effectiveness of our method.
Authors: Shulin Tian, Yufei Wang, Renjie Wan, Wenhan Yang, Alex C. Kot, Bihan Wen
Low-light image enhancement task is essential yet challenging as it is ill-posed intrinsically. Previous arts mainly focus on the low-light images captured in the visible spectrum using pixel-wise loss, which limits the capacity of recovering the brightness, contrast, and texture details due to the small number of income photons. In this work, we propose a novel approach to increase the visibility of images captured under low-light environments by removing the in-camera infrared (IR) cut-off filter, which allows for the capture of more photons and results in improved signal-to-noise ratio due to the inclusion of information from the IR spectrum. To verify the proposed strategy, we collect a paired dataset of low-light images captured without the IR cut-off filter, with corresponding long-exposure reference images with an external filter. The experimental results on the proposed dataset demonstrate the effectiveness of the proposed method, showing better performance quantitatively and qualitatively. The dataset and code are publicly available at https://wyf0912.github.io/ELIEI/
Authors: Ao Shen, Yijie Zhu, Richard Jiang
Marine debris is an important issue for environmental protection, but current methods for locating marine debris are yet limited. In order to achieve higher efficiency and wider applicability in the localization of Marine debris, this study tries to combine the instance segmentation of YOLOv7 with different attention mechanisms and explores the best model. By utilizing a labelled dataset consisting of satellite images containing ocean debris, we examined three attentional models including lightweight coordinate attention, CBAM (combining spatial and channel focus), and bottleneck transformer (based on self-attention). Box detection assessment revealed that CBAM achieved the best outcome (F1 score of 77%) compared to coordinate attention (F1 score of 71%) and YOLOv7/bottleneck transformer (both F1 scores around 66%). Mask evaluation showed CBAM again leading with an F1 score of 73%, whereas coordinate attention and YOLOv7 had comparable performances (around F1 score of 68%/69%) and bottleneck transformer lagged behind at F1 score of 56%. These findings suggest that CBAM offers optimal suitability for detecting marine debris. However, it should be noted that the bottleneck transformer detected some areas missed by manual annotation and displayed better mask precision for larger debris pieces, signifying potentially superior practical performance.
Authors: Zhiyu Zhu, Junhui Hou, Dapeng Oliver Wu
This paper addresses the problem of cross-modal object tracking from RGB videos and event data. Rather than constructing a complex cross-modal fusion network, we explore the great potential of a pre-trained vision Transformer (ViT). Particularly, we delicately investigate plug-and-play training augmentations that encourage the ViT to bridge the vast distribution gap between the two modalities, enabling comprehensive cross-modal information interaction and thus enhancing its ability. Specifically, we propose a mask modeling strategy that randomly masks a specific modality of some tokens to enforce the interaction between tokens from different modalities interacting proactively. To mitigate network oscillations resulting from the masking strategy and further amplify its positive effect, we then theoretically propose an orthogonal high-rank loss to regularize the attention matrix. Extensive experiments demonstrate that our plug-and-play training augmentation techniques can significantly boost state-of-the-art one-stream and twostream trackers to a large extent in terms of both tracking precision and success rate. Our new perspective and findings will potentially bring insights to the field of leveraging powerful pre-trained ViTs to model cross-modal data. The code will be publicly available.
Authors: Amrit Diggavi Seshadri, Alessandra Russo
In this work, following the intuition that adverbs describing scene-sequences are best identified by reasoning over high-level concepts of object-behavior, we propose the design of a new framework that reasons over object-behaviours extracted from raw-video-clips to recognize the clip's corresponding adverb-types. Importantly, while previous works for general scene adverb-recognition assume knowledge of the clips underlying action-types, our method is directly applicable in the more general problem setting where the action-type of a video-clip is unknown. Specifically, we propose a novel pipeline that extracts human-interpretable object-behaviour-facts from raw video clips and propose novel symbolic and transformer based reasoning methods that operate over these extracted facts to identify adverb-types. Experiment results demonstrate that our proposed methods perform favourably against the previous state-of-the-art. Additionally, to support efforts in symbolic video-processing, we release two new datasets of object-behaviour-facts extracted from raw video clips - the MSR-VTT-ASP and ActivityNet-ASP datasets.
Authors: Yuanheng Zhang, Nan Jiang, Zhaoheng Xie, Junying Cao, Yueyang Teng
Accurately annotated ultrasonic images are vital components of a high-quality medical report. Hospitals often have strict guidelines on the types of annotations that should appear on imaging results. However, manually inspecting these images can be a cumbersome task. While a neural network could potentially automate the process, training such a model typically requires a dataset of paired input and target images, which in turn involves significant human labour. This study introduces an automated approach for detecting annotations in images. This is achieved by treating the annotations as noise, creating a self-supervised pretext task and using a model trained under the Noise2Noise scheme to restore the image to a clean state. We tested a variety of model structures on the denoising task against different types of annotation, including body marker annotation, radial line annotation, etc. Our results demonstrate that most models trained under the Noise2Noise scheme outperformed their counterparts trained with noisy-clean data pairs. The costumed U-Net yielded the most optimal outcome on the body marker annotation dataset, with high scores on segmentation precision and reconstruction similarity. We released our code at https://github.com/GrandArth/UltrasonicImage-N2N-Approach.
Authors: Yilan Zhang, Jianqi Chen, Ke Wang, Fengying Xie
Skin image datasets often suffer from imbalanced data distribution, exacerbating the difficulty of computer-aided skin disease diagnosis. Some recent works exploit supervised contrastive learning (SCL) for this long-tailed challenge. Despite achieving significant performance, these SCL-based methods focus more on head classes, yet ignoring the utilization of information in tail classes. In this paper, we propose class-Enhancement Contrastive Learning (ECL), which enriches the information of minority classes and treats different classes equally. For information enhancement, we design a hybrid-proxy model to generate class-dependent proxies and propose a cycle update strategy for parameters optimization. A balanced-hybrid-proxy loss is designed to exploit relations between samples and proxies with different classes treated equally. Taking both "imbalanced data" and "imbalanced diagnosis difficulty" into account, we further present a balanced-weighted cross-entropy loss following curriculum learning schedule. Experimental results on the classification of imbalanced skin lesion data have demonstrated the superiority and effectiveness of our method.
Authors: Quoc Hung Cao, Truong Thanh Hung Nguyen, Vo Thanh Khang Nguyen, Xuan Phong Nguyen
In recent years, artificial intelligence is increasingly being applied widely in many different fields and has a profound and direct impact on human life. Following this is the need to understand the principles of the model making predictions. Since most of the current high-precision models are black boxes, neither the AI scientist nor the end-user deeply understands what's going on inside these models. Therefore, many algorithms are studied for the purpose of explaining AI models, especially those in the problem of image classification in the field of computer vision such as LIME, CAM, GradCAM. However, these algorithms still have limitations such as LIME's long execution time and CAM's confusing interpretation of concreteness and clarity. Therefore, in this paper, we propose a new method called Segmentation - Class Activation Mapping (SeCAM) that combines the advantages of these algorithms above, while at the same time overcoming their disadvantages. We tested this algorithm with various models, including ResNet50, Inception-v3, VGG16 from ImageNet Large Scale Visual Recognition Challenge (ILSVRC) data set. Outstanding results when the algorithm has met all the requirements for a specific explanation in a remarkably concise time.
Authors: Anurag Dhote, Mohammed Javed, David S Doermann
Charts represent an essential source of visual information in documents and facilitate a deep understanding and interpretation of information typically conveyed numerically. In the scientific literature, there are many charts, each with its stylistic differences. Recently the document understanding community has begun to address the problem of automatic chart understanding, which begins with chart classification. In this paper, we present a survey of the current state-of-the-art techniques for chart classification and discuss the available datasets and their supported chart types. We broadly classify these contributions as traditional approaches based on ML, CNN, and Transformers. Furthermore, we carry out an extensive comparative performance analysis of CNN-based and transformer-based approaches on the recently published CHARTINFO UB-UNITECH PMC dataset for the CHART-Infographics competition at ICPR 2022. The data set includes 15 different chart categories, including 22,923 training images and 13,260 test images. We have implemented a vision-based transformer model that produces state-of-the-art results in chart classification.
Authors: Ayush Singh, Yash Bhambhu, Himanshu Buckchash, Deepak K. Gupta, Dilip K. Prasad
Global contexts in images are quite valuable in image-to-image translation problems. Conventional attention-based and graph-based models capture the global context to a large extent, however, these are computationally expensive. Moreover, the existing approaches are limited to only learning the pairwise semantic relation between any two points on the image. In this paper, we present Latent Graph Attention (LGA) a computationally inexpensive (linear to the number of nodes) and stable, modular framework for incorporating the global context in the existing architectures, especially empowering small-scale architectures to give performance closer to large size architectures, thus making the light-weight architectures more useful for edge devices with lower compute power and lower energy needs. LGA propagates information spatially using a network of locally connected graphs, thereby facilitating to construct a semantically coherent relation between any two spatially distant points that also takes into account the influence of the intermediate pixels. Moreover, the depth of the graph network can be used to adapt the extent of contextual spread to the target dataset, thereby being able to explicitly control the added computational cost. To enhance the learning mechanism of LGA, we also introduce a novel contrastive loss term that helps our LGA module to couple well with the original architecture at the expense of minimal additional computational load. We show that incorporating LGA improves the performance on three challenging applications, namely transparent object segmentation, image restoration for dehazing and optical flow estimation.
Authors: Dan Ruta, Gemma Canet Tarrés, Andrew Gilbert, Eli Shechtman, Nicholas Kolkin, John Collomosse
Neural Style Transfer (NST) is the field of study applying neural techniques to modify the artistic appearance of a content image to match the style of a reference style image. Traditionally, NST methods have focused on texture-based image edits, affecting mostly low level information and keeping most image structures the same. However, style-based deformation of the content is desirable for some styles, especially in cases where the style is abstract or the primary concept of the style is in its deformed rendition of some content. With the recent introduction of diffusion models, such as Stable Diffusion, we can access far more powerful image generation techniques, enabling new possibilities. In our work, we propose using this new class of models to perform style transfer while enabling deformable style transfer, an elusive capability in previous models. We show how leveraging the priors of these models can expose new artistic controls at inference time, and we document our findings in exploring this new direction for the field of style transfer.
Authors: Xavier Bou, Aitor Artola, Thibaud Ehret, Gabriele Facciolo, Jean-Michel Morel, Rafael Grompone von Gioi
Detecting relevant changes is a fundamental problem of video surveillance. Because of the high variability of data and the difficulty of properly annotating changes, unsupervised methods dominate the field. Arguably one of the most critical issues to make them practical is to reduce their false alarm rate. In this work, we develop a method-agnostic weakly supervised a-contrario validation process, based on high dimensional statistical modeling of deep features, to reduce the number of false alarms of any change detection algorithm. We also raise the insufficiency of the conventionally used pixel-wise evaluation, as it fails to precisely capture the performance needs of most real applications. For this reason, we complement pixel-wise metrics with object-wise metrics and evaluate the impact of our approach at both pixel and object levels, on six methods and several sequences from different datasets. Experimental results reveal that the proposed a-contrario validation is able to largely reduce the number of false alarms at both pixel and object levels.
Authors: Goluck Konuko, Stéphane Lathuilière, Giuseppe Valenzise
We address the problem of efficiently compressing video for conferencing-type applications. We build on recent approaches based on image animation, which can achieve good reconstruction quality at very low bitrate by representing face motions with a compact set of sparse keypoints. However, these methods encode video in a frame-by-frame fashion, i.e. each frame is reconstructed from a reference frame, which limits the reconstruction quality when the bandwidth is larger. Instead, we propose a predictive coding scheme which uses image animation as a predictor, and codes the residual with respect to the actual target frame. The residuals can be in turn coded in a predictive manner, thus removing efficiently temporal dependencies. Our experiments indicate a significant bitrate gain, in excess of 70% compared to the HEVC video standard and over 30% compared to VVC, on a datasetof talking-head videos
Authors: Tsai Hor Chan, Fernando Julio Cendra, Lan Ma, Guosheng Yin, Lequan Yu
Graph-based methods have been extensively applied to whole-slide histopathology image (WSI) analysis due to the advantage of modeling the spatial relationships among different entities. However, most of the existing methods focus on modeling WSIs with homogeneous graphs (e.g., with homogeneous node type). Despite their successes, these works are incapable of mining the complex structural relations between biological entities (e.g., the diverse interaction among different cell types) in the WSI. We propose a novel heterogeneous graph-based framework to leverage the inter-relationships among different types of nuclei for WSI analysis. Specifically, we formulate the WSI as a heterogeneous graph with "nucleus-type" attribute to each node and a semantic similarity attribute to each edge. We then present a new heterogeneous-graph edge attribute transformer (HEAT) to take advantage of the edge and node heterogeneity during massage aggregating. Further, we design a new pseudo-label-based semantic-consistent pooling mechanism to obtain graph-level features, which can mitigate the over-parameterization issue of conventional cluster-based pooling. Additionally, observing the limitations of existing association-based localization methods, we propose a causal-driven approach attributing the contribution of each node to improve the interpretability of our framework. Extensive experiments on three public TCGA benchmark datasets demonstrate that our framework outperforms the state-of-the-art methods with considerable margins on various tasks. Our codes are available at https://github.com/HKU-MedAI/WSI-HGNN.
Authors: Wei Han, Hui Chen, Min-Yen Kan, Soujanya Poria
Video question--answering is a fundamental task in the field of video understanding. Although current vision--language models (VLMs) equipped with Video Transformers have enabled temporal modeling and yielded superior results, they are at the cost of huge computational power and thus too expensive to deploy in real-time application scenarios. An economical workaround only samples a small portion of frames to represent the main content of that video and tune an image--text model on these sampled frames. Recent video understanding models usually randomly sample a set of frames or clips, regardless of internal correlations between their visual contents, nor their relevance to the problem. We argue that such kinds of aimless sampling may omit the key frames from which the correct answer can be deduced, and the situation gets worse when the sampling sparsity increases, which always happens as the video lengths increase. To mitigate this issue, we propose two frame sampling strategies, namely the most domain frames (MDF) and most implied frames (MIF), to maximally preserve those frames that are most likely vital to the given questions. MDF passively minimizes the risk of key frame omission in a bootstrap manner, while MIS actively searches key frames customized for each video--question pair with the assistance of auxiliary models. The experimental results on three public datasets from three advanced VLMs (CLIP, GIT and All-in-one) demonstrate that our proposed strategies can boost the performance for image--text pretrained models. The source codes pertaining to the method proposed in this paper are publicly available at https://github.com/declare-lab/sas-vqa.
Authors: Truong-Dong Do, Nghe-Nhan Truong, My-Ha Le
Fire is considered one of the most serious threats to human lives which results in a high probability of fatalities. Those severe consequences stem from the heavy smoke emitted from a fire that mostly restricts the visibility of escaping victims and rescuing squad. In such hazardous circumstances, the use of a vision-based human detection system is able to improve the ability to save more lives. To this end, a thermal and infrared imaging fusion strategy based on multiple cameras for human detection in low-visibility scenarios caused by smoke is proposed in this paper. By processing with multiple cameras, vital information can be gathered to generate more useful features for human detection. Firstly, the cameras are calibrated using a Light Heating Chessboard. Afterward, the features extracted from the input images are merged prior to being passed through a lightweight deep neural network to perform the human detection task. The experiments conducted on an NVIDIA Jetson Nano computer demonstrated that the proposed method can process with reasonable speed and can achieve favorable performance with a mAP@0.5 of 95%.
Authors: Boxiang Zhang, Zunran Wang, Yonggen Ling, Yuanyuan Guan, Shenghao Zhang, Wenhui Li
Existing methods of cross-modal domain adaptation for 3D semantic segmentation predict results only via 2D-3D complementarity that is obtained by cross-modal feature matching. However, as lacking supervision in the target domain, the complementarity is not always reliable. The results are not ideal when the domain gap is large. To solve the problem of lacking supervision, we introduce masked modeling into this task and propose a method Mx2M, which utilizes masked cross-modality modeling to reduce the large domain gap. Our Mx2M contains two components. One is the core solution, cross-modal removal and prediction (xMRP), which makes the Mx2M adapt to various scenarios and provides cross-modal self-supervision. The other is a new way of cross-modal feature matching, the dynamic cross-modal filter (DxMF) that ensures the whole method dynamically uses more suitable 2D-3D complementarity. Evaluation of the Mx2M on three DA scenarios, including Day/Night, USA/Singapore, and A2D2/SemanticKITTI, brings large improvements over previous methods on many metrics.
Authors: Aishik Rakshit, Samyak Mehta, Anirban Dasgupta
Optical Character Recognition (OCR) technology finds applications in digitizing books and unstructured documents, along with applications in other domains such as mobility statistics, law enforcement, traffic, security systems, etc. The state-of-the-art methods work well with the OCR with printed text on license plates, shop names, etc. However, applications such as printed textbooks and handwritten texts have limited accuracy with existing techniques. The reason may be attributed to similar-looking characters and variations in handwritten characters. Since these issues are challenging to address with OCR technologies exclusively, we propose a post-processing approach using Natural Language Processing (NLP) tools. This work presents an end-to-end pipeline that first performs OCR on the handwritten or printed text and then improves its accuracy using NLP.
Authors: Vaibhav Vavilala, David Forsyth
We describe a method to parse a complex, cluttered indoor scene into primitives which offer a parsimonious abstraction of scene structure. Our primitives are simple convexes. Our method uses a learned regression procedure to parse a scene into a fixed number of convexes from RGBD input, and can optionally accept segmentations to improve the decomposition. The result is then polished with a descent method which adjusts the convexes to produce a very good fit, and greedily removes superfluous primitives. Because the entire scene is parsed, we can evaluate using traditional depth, normal, and segmentation error metrics. Our evaluation procedure demonstrates that the error from our primitive representation is comparable to that of predicting depth from a single image.
Authors: Jinbao Wang, Guoyang Xie, Yawen Huang, Jiayi Lyu, Feng Zheng, Yefeng Zheng, Yaochu Jin
The problem of how to assess cross-modality medical image synthesis has been largely unexplored. The most used measures like PSNR and SSIM focus on analyzing the structural features but neglect the crucial lesion location and fundamental k-space speciality of medical images. To overcome this problem, we propose a new metric K-CROSS to spur progress on this challenging problem. Specifically, K-CROSS uses a pre-trained multi-modality segmentation network to predict the lesion location, together with a tumor encoder for representing features, such as texture details and brightness intensities. To further reflect the frequency-specific information from the magnetic resonance imaging principles, both k-space features and vision features are obtained and employed in our comprehensive encoders with a frequency reconstruction penalty. The structure-shared encoders are designed and constrained with a similarity loss to capture the intrinsic common structural information for both modalities. As a consequence, the features learned from lesion regions, k-space, and anatomical structures are all captured, which serve as our quality evaluators. We evaluate the performance by constructing a large-scale cross-modality neuroimaging perceptual similarity (NIRPS) dataset with 6,000 radiologist judgments. Extensive experiments demonstrate that the proposed method outperforms other metrics, especially in comparison with the radiologists on NIRPS.
Authors: Max Kaufmann, Daniel Kang, Yi Sun, Steven Basart, Xuwang Yin, Mantas Mazeika, Akul Arora, Adam Dziedzic, Franziska Boenisch, Tom Brown, Jacob Steinhardt, Dan Hendrycks
When considering real-world adversarial settings, defenders are unlikely to have access to the full range of deployment-time adversaries during training, and adversaries are likely to use realistic adversarial distortions that will not be limited to small L_p-constrained perturbations. To narrow in on this discrepancy between research and reality we introduce eighteen novel adversarial attacks, which we use to create ImageNet-UA, a new benchmark for evaluating model robustness against a wide range of unforeseen adversaries. We make use of our benchmark to identify a range of defense strategies which can help overcome this generalization gap, finding a rich space of techniques which can improve unforeseen robustness. We hope the greater variety and realism of ImageNet-UA will make it a useful tool for those working on real-world worst-case robustness, enabling development of more robust defenses which can generalize beyond attacks seen during training.
Authors: Kai Han, Yunhe Wang, Hanting Chen, Xinghao Chen, Jianyuan Guo, Zhenhua Liu, Yehui Tang, An Xiao, Chunjing Xu, Yixing Xu, Zhaohui Yang, Yiman Zhang, Dacheng Tao
Transformer, first applied to the field of natural language processing, is a type of deep neural network mainly based on the self-attention mechanism. Thanks to its strong representation capabilities, researchers are looking at ways to apply transformer to computer vision tasks. In a variety of visual benchmarks, transformer-based models perform similar to or better than other types of networks such as convolutional and recurrent neural networks. Given its high performance and less need for vision-specific inductive bias, transformer is receiving more and more attention from the computer vision community. In this paper, we review these vision transformer models by categorizing them in different tasks and analyzing their advantages and disadvantages. The main categories we explore include the backbone network, high/mid-level vision, low-level vision, and video processing. We also include efficient transformer methods for pushing transformer into real device-based applications. Furthermore, we also take a brief look at the self-attention mechanism in computer vision, as it is the base component in transformer. Toward the end of this paper, we discuss the challenges and provide several further research directions for vision transformers.
Authors: Ruihan Yang, Yibo Yang, Joseph Marino, Stephan Mandt
While recent machine learning research has revealed connections between deep generative models such as VAEs and rate-distortion losses used in learned compression, most of this work has focused on images. In a similar spirit, we view recently proposed neural video coding algorithms through the lens of deep autoregressive and latent variable modeling. We present these codecs as instances of a generalized stochastic temporal autoregressive transform, and propose new avenues for further improvements inspired by normalizing flows and structured priors. We propose several architectures that yield state-of-the-art video compression performance on high-resolution video and discuss their tradeoffs and ablations. In particular, we propose (i) improved temporal autoregressive transforms, (ii) improved entropy models with structured and temporal dependencies, and (iii) variable bitrate versions of our algorithms. Since our improvements are compatible with a large class of existing models, we provide further evidence that the generative modeling viewpoint can advance the neural video coding field.
Authors: James Urquhart Allingham, Florian Wenzel, Zelda E Mariet, Basil Mustafa, Joan Puigcerver, Neil Houlsby, Ghassen Jerfel, Vincent Fortuin, Balaji Lakshminarayanan, Jasper Snoek, Dustin Tran, Carlos Riquelme Ruiz, Rodolphe Jenatton
Machine learning models based on the aggregated outputs of submodels, either at the activation or prediction levels, often exhibit strong performance compared to individual models. We study the interplay of two popular classes of such models: ensembles of neural networks and sparse mixture of experts (sparse MoEs). First, we show that the two approaches have complementary features whose combination is beneficial. This includes a comprehensive evaluation of sparse MoEs in uncertainty related benchmarks. Then, we present Efficient Ensemble of Experts (E$^3$), a scalable and simple ensemble of sparse MoEs that takes the best of both classes of models, while using up to 45% fewer FLOPs than a deep ensemble. Extensive experiments demonstrate the accuracy, log-likelihood, few-shot learning, robustness, and uncertainty improvements of E$^3$ over several challenging vision Transformer-based baselines. E$^3$ not only preserves its efficiency while scaling to models with up to 2.7B parameters, but also provides better predictive performance and uncertainty estimates for larger models.
Authors: Haobo Yuan, Teng Chen, Wei Sui, Jiafeng Xie, Lefei Zhang, Yuan Li, Qian Zhang
Estimating the 3D structure of the drivable surface and surrounding environment is a crucial task for assisted and autonomous driving. It is commonly solved either by using 3D sensors such as LiDAR or directly predicting the depth of points via deep learning. However, the former is expensive, and the latter lacks the use of geometry information for the scene. In this paper, instead of following existing methodologies, we propose Road Planar Parallax Attention Network (RPANet), a new deep neural network for 3D sensing from monocular image sequences based on planar parallax, which takes full advantage of the omnipresent road plane geometry in driving scenes. RPANet takes a pair of images aligned by the homography of the road plane as input and outputs a $\gamma$ map (the ratio of height to depth) for 3D reconstruction. The $\gamma$ map has the potential to construct a two-dimensional transformation between two consecutive frames. It implies planar parallax and can be combined with the road plane serving as a reference to estimate the 3D structure by warping the consecutive frames. Furthermore, we introduce a novel cross-attention module to make the network better perceive the displacements caused by planar parallax. To verify the effectiveness of our method, we sample data from the Waymo Open Dataset and construct annotations related to planar parallax. Comprehensive experiments are conducted on the sampled dataset to demonstrate the 3D reconstruction accuracy of our approach in challenging scenarios.
Authors: Yuqian Chen, Chaoyi Zhang, Tengfei Xue, Yang Song, Nikos Makris, Yogesh Rathi, Weidong Cai, Fan Zhang, Lauren J. O'Donnell
White matter fiber clustering is an important strategy for white matter parcellation, which enables quantitative analysis of brain connections in health and disease. In combination with expert neuroanatomical labeling, data-driven white matter fiber clustering is a powerful tool for creating atlases that can model white matter anatomy across individuals. While widely used fiber clustering approaches have shown good performance using classical unsupervised machine learning techniques, recent advances in deep learning reveal a promising direction toward fast and effective fiber clustering. In this work, we propose a novel deep learning framework for white matter fiber clustering, Deep Fiber Clustering (DFC), which solves the unsupervised clustering problem as a self-supervised learning task with a domain-specific pretext task to predict pairwise fiber distances. This process learns a high-dimensional embedding feature representation for each fiber, regardless of the order of fiber points reconstructed during tractography. We design a novel network architecture that represents input fibers as point clouds and allows the incorporation of additional sources of input information from gray matter parcellation to improve anatomical coherence of clusters. In addition, DFC conducts outlier removal naturally by rejecting fibers with low cluster assignment probability. We evaluate DFC on three independently acquired cohorts, including data from 220 individuals across genders, ages (young and elderly adults), and different health conditions (healthy control and multiple neuropsychiatric disorders). We compare DFC to several state-of-the-art white matter fiber clustering algorithms. Experimental results demonstrate superior performance of DFC in terms of cluster compactness, generalization ability, anatomical coherence, and computational efficiency.
Authors: Kirill Bykov, Mayukh Deb, Dennis Grinwald, Klaus-Robert Müller, Marina M.-C. Höhne
Deep Neural Networks (DNNs) excel at learning complex abstractions within their internal representations. However, the concepts they learn remain opaque, a problem that becomes particularly acute when models unintentionally learn spurious correlations. In this work, we present DORA (Data-agnOstic Representation Analysis), the first data-agnostic framework for analyzing the representational space of DNNs. Central to our framework is the proposed Extreme-Activation (EA) distance measure, which assesses similarities between representations by analyzing their activation patterns on data points that cause the highest level of activation. As spurious correlations often manifest in features of data that are anomalous to the desired task, such as watermarks or artifacts, we demonstrate that internal representations capable of detecting such artifactual concepts can be found by analyzing relationships within neural representations. We validate the EA metric quantitatively, demonstrating its effectiveness both in controlled scenarios and real-world applications. Finally, we provide practical examples from popular Computer Vision models to illustrate that representations identified as outliers using the EA metric often correspond to undesired and spurious concepts.
Authors: Xin Zhang, Ying-Cong Chen
Deep neural networks suffer from significant performance deterioration when there exists distribution shift between deployment and training. Domain Generalization (DG) aims to safely transfer a model to unseen target domains by only relying on a set of source domains. Although various DG approaches have been proposed, a recent study named DomainBed, reveals that most of them do not beat the simple Empirical Risk Minimization (ERM). To this end, we propose a general framework that is orthogonal to existing DG algorithms and could improve their performance consistently. Unlike previous DG works that stake on a static source model to be hopefully a universal one, our proposed AdaODM adaptively modifies the source model at test time for different target domains. Specifically, we create multiple domain-specific classifiers upon a shared domain-generic feature extractor. The feature extractor and classifiers are trained in an adversarial way, where the feature extractor embeds the input samples into a domain-invariant space, and the multiple classifiers capture the distinct decision boundaries that each of them relates to a specific source domain. During testing, distribution differences between target and source domains could be effectively measured by leveraging prediction disagreement among source classifiers. By fine-tuning source models to minimize the disagreement at test time, target domain features are well aligned to the invariant feature space. We verify AdaODM on two popular DG methods, namely ERM and CORAL, and four DG benchmarks, namely VLCS, PACS, OfficeHome, and TerraIncognita. The results show AdaODM stably improves the generalization capacity on unseen domains and achieves state-of-the-art performance.
Authors: Ashwin De Silva, Rahul Ramesh, Carey E. Priebe, Pratik Chaudhari, Joshua T. Vogelstein
We expect the generalization error to improve with more samples from a similar task, and to deteriorate with more samples from an out-of-distribution (OOD) task. In this work, we show a counter-intuitive phenomenon: the generalization error of a task can be a non-monotonic function of the number of OOD samples. As the number of OOD samples increases, the generalization error on the target task improves before deteriorating beyond a threshold. In other words, there is value in training on small amounts of OOD data. We use Fisher's Linear Discriminant on synthetic datasets and deep networks on computer vision benchmarks such as MNIST, CIFAR-10, CINIC-10, PACS and DomainNet to demonstrate and analyze this phenomenon. In the idealistic setting where we know which samples are OOD, we show that these non-monotonic trends can be exploited using an appropriately weighted objective of the target and OOD empirical risk. While its practical utility is limited, this does suggest that if we can detect OOD samples, then there may be ways to benefit from them. When we do not know which samples are OOD, we show how a number of go-to strategies such as data-augmentation, hyper-parameter optimization, and pre-training are not enough to ensure that the target generalization error does not deteriorate with the number of OOD samples in the dataset.
Authors: Qingyun Wang, Manling Li, Hou Pong Chan, Lifu Huang, Julia Hockenmaier, Girish Chowdhary, Heng Ji
Goal-oriented generative script learning aims to generate subsequent steps to reach a particular goal, which is an essential task to assist robots or humans in performing stereotypical activities. An important aspect of this process is the ability to capture historical states visually, which provides detailed information that is not covered by text and will guide subsequent steps. Therefore, we propose a new task, Multimedia Generative Script Learning, to generate subsequent steps by tracking historical states in both text and vision modalities, as well as presenting the first benchmark containing 5,652 tasks and 79,089 multimedia steps. This task is challenging in three aspects: the multimedia challenge of capturing the visual states in images, the induction challenge of performing unseen tasks, and the diversity challenge of covering different information in individual steps. We propose to encode visual state changes through a selective multimedia encoder to address the multimedia challenge, transfer knowledge from previously observed tasks using a retrieval-augmented decoder to overcome the induction challenge, and further present distinct information at each step by optimizing a diversity-oriented contrastive learning objective. We define metrics to evaluate both generation and inductive quality. Experiment results demonstrate that our approach significantly outperforms strong baselines.
Authors: Quang Pham, Chenghao Liu, Steven C. H. Hoi
According to the Complementary Learning Systems (CLS) theory~\cite{mcclelland1995there} in neuroscience, humans do effective \emph{continual learning} through two complementary systems: a fast learning system centered on the hippocampus for rapid learning of the specifics, individual experiences; and a slow learning system located in the neocortex for the gradual acquisition of structured knowledge about the environment. Motivated by this theory, we propose \emph{DualNets} (for Dual Networks), a general continual learning framework comprising a fast learning system for supervised learning of pattern-separated representation from specific tasks and a slow learning system for representation learning of task-agnostic general representation via Self-Supervised Learning (SSL). DualNets can seamlessly incorporate both representation types into a holistic framework to facilitate better continual learning in deep neural networks. Via extensive experiments, we demonstrate the promising results of DualNets on a wide range of continual learning protocols, ranging from the standard offline, task-aware setting to the challenging online, task-free scenario. Notably, on the CTrL~\cite{veniat2020efficient} benchmark that has unrelated tasks with vastly different visual images, DualNets can achieve competitive performance with existing state-of-the-art dynamic architecture strategies~\cite{ostapenko2021continual}. Furthermore, we conduct comprehensive ablation studies to validate DualNets efficacy, robustness, and scalability. Code will be made available at \url{https://github.com/phquang/DualNet}.
Authors: Hu Ding, Wenjie Liu, Mingquan Ye
Many real-world problems can be formulated as the alignment between two geometric patterns. Previously, a great amount of research focus on the alignment of 2D or 3D patterns in the field of computer vision. Recently, the alignment problem in high dimensions finds several novel applications in practice. However, the research is still rather limited in the algorithmic aspect. To the best of our knowledge, most existing approaches are just simple extensions of their counterparts for 2D and 3D cases, and often suffer from the issues such as high computational complexities. In this paper, we propose an effective framework to compress the high dimensional geometric patterns. Any existing alignment method can be applied to the compressed geometric patterns and the time complexity can be significantly reduced. Our idea is inspired by the observation that high dimensional data often has a low intrinsic dimension. Our framework is a ``data-dependent'' approach that has the complexity depending on the intrinsic dimension of the input data. Our experimental results reveal that running the alignment algorithm on compressed patterns can achieve similar qualities, comparing with the results on the original patterns, but the runtimes (including the times cost for compression) are substantially lower.
Authors: Hongyi Wang, Lanfen Lin, Hongjie Hu, Qingqing Chen, Yinhao Li, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen, Ruofeng Tong
High resolution (HR) 3D images are widely used nowadays, such as medical images like Magnetic Resonance Imaging (MRI) and Computed Tomography (CT). However, segmentation of these 3D images remains a challenge due to their high spatial resolution and dimensionality in contrast to currently limited GPU memory. Therefore, most existing 3D image segmentation methods use patch-based models, which have low inference efficiency and ignore global contextual information. To address these problems, we propose a super-resolution (SR) based patch-free 3D image segmentation framework that can realize HR segmentation from a global-wise low-resolution (LR) input. The framework contains two sub-tasks, of which semantic segmentation is the main task and super resolution is an auxiliary task aiding in rebuilding the high frequency information from the LR input. To furthermore balance the information loss with the LR input, we propose a High-Frequency Guidance Module (HGM), and design an efficient selective cropping algorithm to crop an HR patch from the original image as restoration guidance for it. In addition, we also propose a Task-Fusion Module (TFM) to exploit the inter connections between segmentation and SR task, realizing joint optimization of the two tasks. When predicting, only the main segmentation task is needed, while other modules can be removed for acceleration. The experimental results on two different datasets show that our framework has a four times higher inference speed compared to traditional patch-based methods, while its performance also surpasses other patch-based and patch-free models.
Authors: Zhijie Wu, Yuhe Jin, Kwang Moo Yi
We present a novel method to provide efficient and highly detailed reconstructions. Inspired by wavelets, we learn a neural field that decompose the signal both spatially and frequency-wise. We follow the recent grid-based paradigm for spatial decomposition, but unlike existing work, encourage specific frequencies to be stored in each grid via Fourier features encodings. We then apply a multi-layer perceptron with sine activations, taking these Fourier encoded features in at appropriate layers so that higher-frequency components are accumulated on top of lower-frequency components sequentially, which we sum up to form the final output. We demonstrate that our method outperforms the state of the art regarding model compactness and convergence speed on multiple tasks: 2D image fitting, 3D shape reconstruction, and neural radiance fields. Our code is available at https://github.com/ubc-vision/NFFB.
Authors: Yunshan Zhong, Lizhou You, Yuxin Zhang, Fei Chao, Yonghong Tian, Rongrong Ji
Most shadow removal methods rely on the invasion of training images associated with laborious and lavish shadow region annotations, leading to the increasing popularity of shadow image synthesis. However, the poor performance also stems from these synthesized images since they are often shadow-inauthentic and details-impaired. In this paper, we present a novel generation framework, referred to as HQSS, for high-quality pseudo shadow image synthesis. The given image is first decoupled into a shadow region identity and a non-shadow region identity. HQSS employs a shadow feature encoder and a generator to synthesize pseudo images. Specifically, the encoder extracts the shadow feature of a region identity which is then paired with another region identity to serve as the generator input to synthesize a pseudo image. The pseudo image is expected to have the shadow feature as its input shadow feature and as well as a real-like image detail as its input region identity. To fulfill this goal, we design three learning objectives. When the shadow feature and input region identity are from the same region identity, we propose a self-reconstruction loss that guides the generator to reconstruct an identical pseudo image as its input. When the shadow feature and input region identity are from different identities, we introduce an inter-reconstruction loss and a cycle-reconstruction loss to make sure that shadow characteristics and detail information can be well retained in the synthesized images. Our HQSS is observed to outperform the state-of-the-art methods on ISTD dataset, Video Shadow Removal dataset, and SRD dataset. The code is available at https://github.com/zysxmu/HQSS.
Authors: Olivier Pradelle, Raphaelle Chaine, David Wendland, Julie Digne
Scene understanding has made tremendous progress over the past few years, as data acquisition systems are now providing an increasing amount of data of various modalities (point cloud, depth, RGB...). However, this improvement comes at a large cost on computation resources and data annotation requirements. To analyze geometric information and images jointly, many approaches rely on both a 2D loss and 3D loss, requiring not only 2D per pixel-labels but also 3D per-point labels. However, obtaining a 3D groundtruth is challenging, time-consuming and error-prone. In this paper, we show that image segmentation can benefit from 3D geometric information without requiring a 3D groundtruth, by training the geometric feature extraction and the 2D segmentation network jointly, in an end-to-end fashion, using only the 2D segmentation loss. Our method starts by extracting a map of 3D features directly from a provided point cloud by using a lightweight 3D neural network. The 3D feature map, merged with the RGB image, is then used as an input to a classical image segmentation network. Our method can be applied to many 2D segmentation networks, improving significantly their performance with only a marginal network weight increase and light input dataset requirements, since no 3D groundtruth is required.
Authors: Munan Ning, Donghuan Lu, Yujia Xie, Dongdong Chen, Dong Wei, Yefeng Zheng, Yonghong Tian, Shuicheng Yan, Li Yuan
Unsupervised domain adaption has been widely adopted in tasks with scarce annotated data. Unfortunately, mapping the target-domain distribution to the source-domain unconditionally may distort the essential structural information of the target-domain data, leading to inferior performance. To address this issue, we firstly propose to introduce active sample selection to assist domain adaptation regarding the semantic segmentation task. By innovatively adopting multiple anchors instead of a single centroid, both source and target domains can be better characterized as multimodal distributions, in which way more complementary and informative samples are selected from the target domain. With only a little workload to manually annotate these active samples, the distortion of the target-domain distribution can be effectively alleviated, achieving a large performance gain. In addition, a powerful semi-supervised domain adaptation strategy is proposed to alleviate the long-tail distribution problem and further improve the segmentation performance. Extensive experiments are conducted on public datasets, and the results demonstrate that the proposed approach outperforms state-of-the-art methods by large margins and achieves similar performance to the fully-supervised upperbound, i.e., 71.4% mIoU on GTA5 and 71.8% mIoU on SYNTHIA. The effectiveness of each component is also verified by thorough ablation studies.
Authors: Guoyang Xie, Jinbao Wang, Jiaqi Liu, Jiayi Lyu, Yong Liu, Chengjie Wang, Feng Zheng, Yaochu Jin
Image anomaly detection (IAD) is an emerging and vital computer vision task in industrial manufacturing (IM). Recently many advanced algorithms have been published, but their performance deviates greatly. We realize that the lack of actual IM settings most probably hinders the development and usage of these methods in real-world applications. As far as we know, IAD methods are not evaluated systematically. As a result, this makes it difficult for researchers to analyze them because they are designed for different or special cases. To solve this problem, we first propose a uniform IM setting to assess how well these algorithms perform, which includes several aspects, i.e., various levels of supervision (unsupervised vs. semi-supervised), few-shot learning, continual learning, noisy labels, memory usage, and inference speed. Moreover, we skillfully build a comprehensive image anomaly detection benchmark (IM-IAD) that includes 16 algorithms on 7 mainstream datasets with uniform settings. Our extensive experiments (17,017 in total) provide in-depth insights for IAD algorithm redesign or selection under the IM setting. Next, the proposed benchmark IM-IAD gives challenges as well as directions for the future. To foster reproducibility and accessibility, the source code of IM-IAD is uploaded on the website, https://github.com/M-3LAB/IM-IAD.
Authors: Kun Wang, Zi Wang, Zhang Li, Ang Su, Xichao Teng, Minhao Liu, Qifeng Yu
Oriented object detection is one of the most fundamental and challenging tasks in remote sensing, aiming at locating the oriented objects of numerous predefined object categories. Recently, deep learning based methods have achieved remarkable performance in detecting oriented objects in optical remote sensing imagery. However, a thorough review of the literature in remote sensing has not yet emerged. Therefore, we give a comprehensive survey of recent advances and cover many aspects of oriented object detection, including problem definition, commonly used datasets, evaluation protocols, detection frameworks, oriented object representations, and feature representations. Besides, the state-of-the-art methods are analyzed and discussed. We finally discuss future research directions to put forward some useful research guidance. We believe that this survey shall be valuable to researchers across academia and industry
Authors: Toan Nguyen, Minh Nhat Vu, An Vuong, Dzung Nguyen, Thieu Vo, Ngan Le, Anh Nguyen
Affordance detection is a challenging problem with a wide variety of robotic applications. Traditional affordance detection methods are limited to a predefined set of affordance labels, hence potentially restricting the adaptability of intelligent robots in complex and dynamic environments. In this paper, we present the Open-Vocabulary Affordance Detection (OpenAD) method, which is capable of detecting an unbounded number of affordances in 3D point clouds. By simultaneously learning the affordance text and the point feature, OpenAD successfully exploits the semantic relationships between affordances. Therefore, our proposed method enables zero-shot detection and can be able to detect previously unseen affordances without a single annotation example. Intensive experimental results show that OpenAD works effectively on a wide range of affordance detection setups and outperforms other baselines by a large margin. Additionally, we demonstrate the practicality of the proposed OpenAD in real-world robotic applications with a fast inference speed (~100ms). Our project is available at https://openad2023.github.io.
Authors: Jiahui Fu, Yilun Du, Kurran Singh, Joshua B. Tenenbaum, John J. Leonard
We present NeuSE, a novel Neural SE(3)-Equivariant Embedding for objects, and illustrate how it supports object SLAM for consistent spatial understanding with long-term scene changes. NeuSE is a set of latent object embeddings created from partial object observations. It serves as a compact point cloud surrogate for complete object models, encoding full shape information while transforming SE(3)-equivariantly in tandem with the object in the physical world. With NeuSE, relative frame transforms can be directly derived from inferred latent codes. Our proposed SLAM paradigm, using NeuSE for object shape and pose characterization, can operate independently or in conjunction with typical SLAM systems. It directly infers SE(3) camera pose constraints that are compatible with general SLAM pose graph optimization, while also maintaining a lightweight object-centric map that adapts to real-world changes. Our approach is evaluated on synthetic and real-world sequences featuring changed objects and shows improved localization accuracy and change-aware mapping capability, when working either standalone or jointly with a common SLAM pipeline.
Authors: Paul Friedrich, Julia Wolleb, Florentin Bieder, Florian M. Thieringer, Philippe C. Cattin
Advances in 3D printing of biocompatible materials make patient-specific implants increasingly popular. The design of these implants is, however, still a tedious and largely manual process. Existing approaches to automate implant generation are mainly based on 3D U-Net architectures on downsampled or patch-wise data, which can result in a loss of detail or contextual information. Following the recent success of Diffusion Probabilistic Models, we propose a novel approach for implant generation based on a combination of 3D point cloud diffusion models and voxelization networks. Due to the stochastic sampling process in our diffusion model, we can propose an ensemble of different implants per defect, from which the physicians can choose the most suitable one. We evaluate our method on the SkullBreak and SkullFix datasets, generating high-quality implants and achieving competitive evaluation scores.
Authors: Jindong Jiang, Fei Deng, Gautam Singh, Sungjin Ahn
The recent success of transformer-based image generative models in object-centric learning highlights the importance of powerful image generators for handling complex scenes. However, despite the high expressiveness of diffusion models in image generation, their integration into object-centric learning remains largely unexplored in this domain. In this paper, we explore the feasibility and potential of integrating diffusion models into object-centric learning and investigate the pros and cons of this approach. We introduce Latent Slot Diffusion (LSD), a novel model that serves dual purposes: it is the first object-centric learning model to replace conventional slot decoders with a latent diffusion model conditioned on object slots, and it is also the first unsupervised compositional conditional diffusion model that operates without the need for supervised annotations like text. Through experiments on various object-centric tasks, including the first application of the FFHQ dataset in this field, we demonstrate that LSD significantly outperforms state-of-the-art transformer-based decoders, particularly in more complex scenes, and exhibits superior unsupervised compositional generation quality. Project page is available at $\href{https://latentslotdiffusion.github.io}{here}$
Authors: Changdae Oh, Hyeji Hwang, Hee-young Lee, YongTaek Lim, Geunyoung Jung, Jiyoung Jung, Hosik Choi, Kyungwoo Song
With the surge of large-scale pre-trained models (PTMs), fine-tuning these models to numerous downstream tasks becomes a crucial problem. Consequently, parameter efficient transfer learning (PETL) of large models has grasped huge attention. While recent PETL methods showcase impressive performance, they rely on optimistic assumptions: 1) the entire parameter set of a PTM is available, and 2) a sufficiently large memory capacity for the fine-tuning is equipped. However, in most real-world applications, PTMs are served as a black-box API or proprietary software without explicit parameter accessibility. Besides, it is hard to meet a large memory requirement for modern PTMs. In this work, we propose black-box visual prompting (BlackVIP), which efficiently adapts the PTMs without knowledge about model architectures and parameters. BlackVIP has two components; 1) Coordinator and 2) simultaneous perturbation stochastic approximation with gradient correction (SPSA-GC). The Coordinator designs input-dependent image-shaped visual prompts, which improves few-shot adaptation and robustness on distribution/location shift. SPSA-GC efficiently estimates the gradient of a target model to update Coordinator. Extensive experiments on 16 datasets demonstrate that BlackVIP enables robust adaptation to diverse domains without accessing PTMs' parameters, with minimal memory requirements. Code: \url{https://github.com/changdaeoh/BlackVIP}
Authors: Kun Su, Kaizhi Qian, Eli Shlizerman, Antonio Torralba, Chuang Gan
Modeling sounds emitted from physical object interactions is critical for immersive perceptual experiences in real and virtual worlds. Traditional methods of impact sound synthesis use physics simulation to obtain a set of physics parameters that could represent and synthesize the sound. However, they require fine details of both the object geometries and impact locations, which are rarely available in the real world and can not be applied to synthesize impact sounds from common videos. On the other hand, existing video-driven deep learning-based approaches could only capture the weak correspondence between visual content and impact sounds since they lack of physics knowledge. In this work, we propose a physics-driven diffusion model that can synthesize high-fidelity impact sound for a silent video clip. In addition to the video content, we propose to use additional physics priors to guide the impact sound synthesis procedure. The physics priors include both physics parameters that are directly estimated from noisy real-world impact sound examples without sophisticated setup and learned residual parameters that interpret the sound environment via neural networks. We further implement a novel diffusion model with specific training and inference strategies to combine physics priors and visual information for impact sound synthesis. Experimental results show that our model outperforms several existing systems in generating realistic impact sounds. More importantly, the physics-based representations are fully interpretable and transparent, thus enabling us to perform sound editing flexibly.
Authors: Prakash Chandra Chhipa, Muskaan Chopra, Gopal Mengi, Varun Gupta, Richa Upadhyay, Meenakshi Subhash Chippa, Kanjar De, Rajkumar Saini, Seiichi Uchida, Marcus Liwicki
This work investigates the unexplored usability of self-supervised representation learning in the direction of functional knowledge transfer. In this work, functional knowledge transfer is achieved by joint optimization of self-supervised learning pseudo task and supervised learning task, improving supervised learning task performance. Recent progress in self-supervised learning uses a large volume of data, which becomes a constraint for its applications on small-scale datasets. This work shares a simple yet effective joint training framework that reinforces human-supervised task learning by learning self-supervised representations just-in-time and vice versa. Experiments on three public datasets from different visual domains, Intel Image, CIFAR, and APTOS, reveal a consistent track of performance improvements on classification tasks during joint optimization. Qualitative analysis also supports the robustness of learnt representations. Source code and trained models are available on GitHub.
Authors: Marcus Nordström, Henrik Hult, Atsuto Maki
This work presents a study on label noise in medical image segmentation by considering a noise model based on Gaussian field deformations. Such noise is of interest because it yields realistic looking segmentations and because it is unbiased in the sense that the expected deformation is the identity mapping. Efficient methods for sampling and closed form solutions for the marginal probabilities are provided. Moreover, theoretically optimal solutions to the loss functions cross-entropy and soft-Dice are studied and it is shown how they diverge as the level of noise increases. Based on recent work on loss function characterization, it is shown that optimal solutions to soft-Dice can be recovered by thresholding solutions to cross-entropy with a particular a priori unknown threshold that efficiently can be computed. This raises the question whether the decrease in performance seen when using cross-entropy as compared to soft-Dice is caused by using the wrong threshold. The hypothesis is validated in 5-fold studies on three organ segmentation problems from the TotalSegmentor data set, using 4 different strengths of noise. The results show that changing the threshold leads the performance of cross-entropy to go from systematically worse than soft-Dice to similar or better results than soft-Dice.
Authors: Tingting Liu, Yuan Liu, Chuncheng Zhang, Yuan Liyin, Xiubao Sui, Qian Chen
Since the number of incident energies is limited, it is difficult to directly acquire hyperspectral images (HSI) with high spatial resolution. Considering the high dimensionality and correlation of HSI, super-resolution (SR) of HSI remains a challenge in the absence of auxiliary high-resolution images. Furthermore, it is very important to extract the spatial features effectively and make full use of the spectral information. This paper proposes a novel HSI super-resolution algorithm, termed dual-domain network based on hybrid convolution (SRDNet). Specifically, a dual-domain network is designed to fully exploit the spatial-spectral and frequency information among the hyper-spectral data. To capture inter-spectral self-similarity, a self-attention learning mechanism (HSL) is devised in the spatial domain. Meanwhile the pyramid structure is applied to increase the acceptance field of attention, which further reinforces the feature representation ability of the network. Moreover, to further improve the perceptual quality of HSI, a frequency loss(HFL) is introduced to optimize the model in the frequency domain. The dynamic weighting mechanism drives the network to gradually refine the generated frequency and excessive smoothing caused by spatial loss. Finally, In order to better fully obtain the mapping relationship between high-resolution space and low-resolution space, a hybrid module of 2D and 3D units with progressive upsampling strategy is utilized in our method. Experiments on a widely used benchmark dataset illustrate that the proposed SRDNet method enhances the texture information of HSI and is superior to state-of-the-art methods.
Authors: Hanyu Cai, Ni Ou, Junzheng Wang
This paper presents a novel visual-LiDAR odometry and mapping method with low-drift characteristics. The proposed method is based on two popular approaches, ORB-SLAM and A-LOAM, with monocular scale correction and visual-bootstrapped LiDAR poses initialization modifications. The scale corrector calculates the proportion between the depth of image keypoints recovered by triangulation and that provided by LiDAR, using an outlier rejection process for accuracy improvement. Concerning LiDAR poses initialization, the visual odometry approach gives the initial guesses of LiDAR motions for better performance. This methodology is not only applicable to high-resolution LiDAR but can also adapt to low-resolution LiDAR. To evaluate the proposed SLAM system's robustness and accuracy, we conducted experiments on the KITTI Odometry and S3E datasets. Experimental results illustrate that our method significantly outperforms standalone ORB-SLAM2 and A-LOAM. Furthermore, regarding the accuracy of visual odometry with scale correction, our method performs similarly to the stereo-mode ORB-SLAM2.
Authors: Rui Zhang, Luziwei Leng, Kaiwei Che, Hu Zhang, Jie Cheng, Qinghai Guo, Jiangxing Liao, Ran Cheng
Leveraging the low-power, event-driven computation and the inherent temporal dynamics, spiking neural networks (SNNs) are potentially ideal solutions for processing dynamic and asynchronous signals from event-based sensors. However, due to the challenges in training and the restrictions in architectural design, there are limited examples of competitive SNNs in the realm of event-based dense prediction when compared to artificial neural networks (ANNs). In this paper, we present an efficient spiking encoder-decoder network designed for large-scale event-based semantic segmentation tasks. This is achieved by optimizing the encoder using a hierarchical search method. To enhance learning from dynamic event streams, we harness the inherent adaptive threshold of spiking neurons to modulate network activation. Moreover, we introduce a dual-path Spiking Spatially-Adaptive Modulation (SSAM) block, specifically designed to enhance the representation of sparse events, thereby considerably improving network performance. Our proposed network achieves a 72.57% mean intersection over union (MIoU) on the DDD17 dataset and a 57.22% MIoU on the recently introduced, larger DSEC-Semantic dataset. This performance surpasses the current state-of-the-art ANNs by 4%, whilst consuming significantly less computational resources. To the best of our knowledge, this is the first study demonstrating SNNs outperforming ANNs in demanding event-based semantic segmentation tasks, thereby establishing the vast potential of SNNs in the field of event-based vision. Our source code will be made publicly accessible.
Authors: Peng Dai, Yinda Zhang, Xin Yu, Xiaoyang Lyu, Xiaojuan Qi
Rendering novel view images is highly desirable for many applications. Despite recent progress, it remains challenging to render high-fidelity and view-consistent novel views of large-scale scenes from in-the-wild images with inevitable artifacts (e.g., motion blur). To this end, we develop a hybrid neural rendering model that makes image-based representation and neural 3D representation join forces to render high-quality, view-consistent images. Besides, images captured in the wild inevitably contain artifacts, such as motion blur, which deteriorates the quality of rendered images. Accordingly, we propose strategies to simulate blur effects on the rendered images to mitigate the negative influence of blurriness images and reduce their importance during training based on precomputed quality-aware weights. Extensive experiments on real and synthetic data demonstrate our model surpasses state-of-the-art point-based methods for novel view synthesis. The code is available at https://daipengwa.github.io/Hybrid-Rendering-ProjectPage.
Authors: Jon Crall
The performance of a binary classifier is described by a confusion matrix with four entries: the number of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN).
The Matthew's Correlation Coefficient (MCC), F1, and Fowlkes--Mallows (FM) scores are scalars that summarize a confusion matrix. Both the F1 and FM scores are based on only three of the four entries in the confusion matrix (they ignore TN). In contrast, the MCC takes into account all four entries of the confusion matrix and thus can be seen as providing a more representative picture.
However, in object detection problems, measuring the number of true negatives is so large it is often intractable. Thus we ask, what happens to the MCC as the number of true negatives approaches infinity? This paper provides insight into the relationship between the MCC and FM score by proving that the FM-measure is equal to the limit of the MCC as the number of true negatives approaches infinity.
Authors: Xinyang Huang, Chuang Zhu, Wenkai Chen
In semi-supervised domain adaptation (SSDA), a few labeled target samples of each class help the model to transfer knowledge representation from the fully labeled source domain to the target domain. Many existing methods ignore the benefits of making full use of the labeled target samples from multi-level. To make better use of this additional data, we propose a novel Prototype-based Multi-level Learning (ProML) framework to better tap the potential of labeled target samples. To achieve intra-domain adaptation, we first introduce a pseudo-label aggregation based on the intra-domain optimal transport to help the model align the feature distribution of unlabeled target samples and the prototype. At the inter-domain level, we propose a cross-domain alignment loss to help the model use the target prototype for cross-domain knowledge transfer. We further propose a dual consistency based on prototype similarity and linear classifier to promote discriminative learning of compact target feature representation at the batch level. Extensive experiments on three datasets, including DomainNet, VisDA2017, and Office-Home demonstrate that our proposed method achieves state-of-the-art performance in SSDA.
Authors: Jung Hwan Heo, Seyedarmin Azizi, Arash Fayyazi, Massoud Pedram
Transfer learning has become a popular task adaptation method in the era of foundation models. However, many foundation models require large storage and computing resources, which makes off-the-shelf deployment impractical. Post-training compression techniques such as pruning and quantization can help lower deployment costs. Unfortunately, the resulting performance degradation limits the usability and benefits of such techniques. To close this performance gap, we propose CrAFT, a simple fine-tuning framework that enables effective post-training network compression. In CrAFT, users simply employ the default fine-tuning schedule along with sharpness minimization objective, simultaneously facilitating task adaptation and compression-friendliness. Contrary to the conventional sharpness minimization techniques, which are applied during pretraining, the CrAFT approach adds negligible training overhead as fine-tuning is done in under a couple of minutes or hours with a single GPU. The effectiveness of CrAFT, which is a general-purpose tool that can significantly boost one-shot pruning and post-training quantization, is demonstrated on both convolution-based and attention-based vision foundation models on a variety of target tasks. The code will be made publicly available.
Authors: Teerapong Panboonyuen, Naphat Nithisopa, Panin Pienroj, Laphonchai Jirachuphun, Chaiwasut Watthanasirikrit, Naruepon Pornwiriyakul
Evaluating car damages from misfortune is critical to the car insurance industry. However, the accuracy is still insufficient for real-world applications since the deep learning network is not designed for car damage images as inputs, and its segmented masks are still very coarse. This paper presents MARS (Mask Attention Refinement with Sequential quadtree nodes) for car damage instance segmentation. Our MARS represents self-attention mechanisms to draw global dependencies between the sequential quadtree nodes layer and quadtree transformer to recalibrate channel weights and predict highly accurate instance masks. Our extensive experiments demonstrate that MARS outperforms state-of-the-art (SOTA) instance segmentation methods on three popular benchmarks such as Mask R-CNN [9], PointRend [13], and Mask Transfiner [12], by a large margin of +1.3 maskAP-based R50-FPN backbone and +2.3 maskAP-based R101-FPN backbone on Thai car-damage dataset. Our demos are available at https://github.com/kaopanboonyuen/MARS.
Authors: Eden Dolev, Alaa Awad, Denisa Roberts, Zahra Ebrahimzadeh, Marcin Mejran, Vaibhav Malpani, Mahir Yavuz
In this article, we present our approach to single-modality visual representation learning. Understanding visual representations of product content is vital for recommendations, search, and advertising applications in e-commerce. We detail and contrast techniques used to fine-tune large-scale visual representation learning models in an efficient manner under low-resource settings, including several pretrained backbone architectures, both in the convolutional neural network as well as the vision transformer family. We highlight the challenges for e-commerce applications at-scale and highlight the efforts to more efficiently train, evaluate, and serve visual representations. We present ablation studies evaluating the representation offline performance for several downstream tasks, including our visually similar ad recommendations. To this end, we present a novel text-to-image generative offline evaluation method for visually similar recommendation systems. Finally, we include online results from deployed machine learning systems in production at Etsy.
Authors: Anahita Fathi Kazerooni, Nastaran Khalili, Xinyang Liu, Debanjan Haldar, Zhifan Jiang, Syed Muhammed Anwar, Jake Albrecht, Maruf Adewole, Udunna Anazodo, Hannah Anderson, Sina Bagheri, Ujjwal Baid, Timothy Bergquist, Austin J. Borja, Evan Calabrese, Verena Chung, Gian-Marco Conte, Farouk Dako, James Eddy, Ivan Ezhov, Ariana Familiar, Keyvan Farahani, Shuvanjan Haldar, Juan Eugenio Iglesias, Anastasia Janas, Elaine Johansen, Blaise V Jones, Florian Kofler, Dominic LaBella, Hollie Anne Lai, Koen Van Leemput, Hongwei Bran Li, Nazanin Maleki, Aaron S McAllister, Zeke Meier, Bjoern Menze, Ahmed W Moawad, Khanak K Nandolia, Julija Pavaine, Marie Piraud, Tina Poussaint, Sanjay P Prabhu, Zachary Reitman, Andres Rodriguez, Jeffrey D Rudie, Ibraheem Salman Shaikh, Lubdha M. Shah, Nakul Sheth, Russel Taki Shinohara, et al. (23 additional authors not shown)
Pediatric tumors of the central nervous system are the most common cause of cancer-related death in children. The five-year survival rate for high-grade gliomas in children is less than 20\%. Due to their rarity, the diagnosis of these entities is often delayed, their treatment is mainly based on historic treatment concepts, and clinical trials require multi-institutional collaborations. The MICCAI Brain Tumor Segmentation (BraTS) Challenge is a landmark community benchmark event with a successful history of 12 years of resource creation for the segmentation and analysis of adult glioma. Here we present the CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs 2023 challenge, which represents the first BraTS challenge focused on pediatric brain tumors with data acquired across multiple international consortia dedicated to pediatric neuro-oncology and clinical trials. The BraTS-PEDs 2023 challenge focuses on benchmarking the development of volumentric segmentation algorithms for pediatric brain glioma through standardized quantitative performance evaluation metrics utilized across the BraTS 2023 cluster of challenges. Models gaining knowledge from the BraTS-PEDs multi-parametric structural MRI (mpMRI) training data will be evaluated on separate validation and unseen test mpMRI dataof high-grade pediatric glioma. The CBTN-CONNECT-DIPGR-ASNR-MICCAI BraTS-PEDs 2023 challenge brings together clinicians and AI/imaging scientists to lead to faster development of automated segmentation techniques that could benefit clinical trials, and ultimately the care of children with brain tumors.
Authors: Jiakang Yuan, Bo Zhang, Xiangchao Yan, Tao Chen, Botian Shi, Yikang Li, Yu Qiao
It is a long-term vision for Autonomous Driving (AD) community that the perception models can learn from a large-scale point cloud dataset, to obtain unified representations that can achieve promising results on different tasks or benchmarks. Previous works mainly focus on the self-supervised pre-training pipeline, meaning that they perform the pre-training and fine-tuning on the same benchmark, which is difficult to attain the performance scalability and cross-dataset application for the pre-training checkpoint. In this paper, for the first time, we are committed to building a large-scale pre-training point-cloud dataset with diverse data distribution, and meanwhile learning generalizable representations from such a diverse pre-training dataset. We formulate the point-cloud pre-training task as a semi-supervised problem, which leverages the few-shot labeled and massive unlabeled point-cloud data to generate the unified backbone representations that can be directly applied to many baseline models and benchmarks, decoupling the AD-related pre-training process and downstream fine-tuning task. During the period of backbone pre-training, by enhancing the scene- and instance-level distribution diversity and exploiting the backbone's ability to learn from unknown instances, we achieve significant performance gains on a series of downstream perception benchmarks including Waymo, nuScenes, and KITTI, under different baseline models like PV-RCNN++, SECOND, CenterPoint.
Authors: Kun Zhou, Wenbo Li, Nianjuan Jiang, Xiaoguang Han, Jiangbo Lu
Neural radiance fields (NeRF) have shown great success in novel view synthesis. However, recovering high-quality details from real-world scenes is still challenging for the existing NeRF-based approaches, due to the potential imperfect calibration information and scene representation inaccuracy. Even with high-quality training frames, the synthetic novel views produced by NeRF models still suffer from notable rendering artifacts, such as noise and blur. To address this, we propose NeRFLiX, a general NeRF-agnostic restorer paradigm that learns a degradation-driven inter-viewpoint mixer. Specially, we design a NeRF-style degradation modeling approach and construct large-scale training data, enabling the possibility of effectively removing NeRF-native rendering artifacts for deep neural networks. Moreover, beyond the degradation removal, we propose an inter-viewpoint aggregation framework that fuses highly related high-quality training images, pushing the performance of cutting-edge NeRF models to entirely new levels and producing highly photo-realistic synthetic views. Based on this paradigm, we further present NeRFLiX++ with a stronger two-stage NeRF degradation simulator and a faster inter-viewpoint mixer, achieving superior performance with significantly improved computational efficiency. Notably, NeRFLiX++ is capable of restoring photo-realistic ultra-high-resolution outputs from noisy low-resolution NeRF-rendered views. Extensive experiments demonstrate the excellent restoration ability of NeRFLiX++ on various novel view synthesis benchmarks.
Authors: Heqin Zhu, Quan Quan, Qingsong Yao, Zaiyi Liu, S. kevin Zhou
One-shot medical landmark detection gains much attention and achieves great success for its label-efficient training process. However, existing one-shot learning methods are highly specialized in a single domain and suffer domain preference heavily in the situation of multi-domain unlabeled data. Moreover, one-shot learning is not robust that it faces performance drop when annotating a sub-optimal image. To tackle these issues, we resort to developing a domain-adaptive one-shot landmark detection framework for handling multi-domain medical images, named Universal One-shot Detection (UOD). UOD consists of two stages and two corresponding universal models which are designed as combinations of domain-specific modules and domain-shared modules. In the first stage, a domain-adaptive convolution model is self-supervised learned to generate pseudo landmark labels. In the second stage, we design a domain-adaptive transformer to eliminate domain preference and build the global context for multi-domain data. Even though only one annotated sample from each domain is available for training, the domain-shared modules help UOD aggregate all one-shot samples to detect more robust and accurate landmarks. We investigated both qualitatively and quantitatively the proposed UOD on three widely-used public X-ray datasets in different anatomical domains (i.e., head, hand, chest) and obtained state-of-the-art performances in each domain.
Authors: Makito Kobayashi, Chen-Chieh Liao, Keito Inoue, Sentaro Yojima, Masafumi Takahashi
In this work, we proposed a new style-diverse dataset for the domain of motion style transfer. The motion dataset uses an industrial-standard human bone structure and thus is industry-ready to be plugged into 3D characters for many projects. We claim the challenges in motion style transfer and encourage future work in this domain by releasing the proposed motion dataset both to the public and the market. We conduct a comprehensive study on motion style transfer in the experiment using the state-of-the-art method, and the results show the proposed dataset's validity for the motion style transfer task.
Authors: Shivam Mehta, Siyang Wang, Simon Alexanderson, Jonas Beskow, Éva Székely, Gustav Eje Henter
With read-aloud speech synthesis achieving high naturalness scores, there is a growing research interest in synthesising spontaneous speech. However, human spontaneous face-to-face conversation has both spoken and non-verbal aspects (here, co-speech gestures). Only recently has research begun to explore the benefits of jointly synthesising these two modalities in a single system. The previous state of the art used non-probabilistic methods, which fail to capture the variability of human speech and motion, and risk producing oversmoothing artefacts and sub-optimal synthesis quality. We present the first diffusion-based probabilistic model, called Diff-TTSG, that jointly learns to synthesise speech and gestures together. Our method can be trained on small datasets from scratch. Furthermore, we describe a set of careful uni- and multi-modal subjective tests for evaluating integrated speech and gesture synthesis systems, and use them to validate our proposed approach. For synthesised examples please see https://shivammehta25.github.io/Diff-TTSG
Authors: Wolfgang Paier, Anna Hilsmann, Peter Eisert
This paper presents a novel approach for text/speech-driven animation of a photo-realistic head model based on blend-shape geometry, dynamic textures, and neural rendering. Training a VAE for geometry and texture yields a parametric model for accurate capturing and realistic synthesis of facial expressions from a latent feature vector. Our animation method is based on a conditional CNN that transforms text or speech into a sequence of animation parameters. In contrast to previous approaches, our animation model learns disentangling/synthesizing different acting-styles in an unsupervised manner, requiring only phonetic labels that describe the content of training sequences. For realistic real-time rendering, we train a U-Net that refines rasterization-based renderings by computing improved pixel colors and a foreground matte. We compare our framework qualitatively/quantitatively against recent methods for head modeling as well as facial animation and evaluate the perceived rendering/animation quality in a user-study, which indicates large improvements compared to state-of-the-art approaches
Authors: Duy M. H. Nguyen, Hoang Nguyen, Nghiem T. Diep, Tan N. Pham, Tri Cao, Binh T. Nguyen, Paul Swoboda, Nhat Ho, Shadi Albarqouni, Pengtao Xie, Daniel Sonntag, Mathias Niepert
Obtaining large pre-trained models that can be fine-tuned to new tasks with limited annotated samples has remained an open challenge for medical imaging data. While pre-trained deep networks on ImageNet and vision-language foundation models trained on web-scale data are prevailing approaches, their effectiveness on medical tasks is limited due to the significant domain shift between natural and medical images. To bridge this gap, we introduce LVM-Med, the first family of deep networks trained on large-scale medical datasets. We have collected approximately 1.3 million medical images from 55 publicly available datasets, covering a large number of organs and modalities such as CT, MRI, X-ray, and Ultrasound. We benchmark several state-of-the-art self-supervised algorithms on this dataset and propose a novel self-supervised contrastive learning algorithm using a graph-matching formulation. The proposed approach makes three contributions: (i) it integrates prior pair-wise image similarity metrics based on local and global information; (ii) it captures the structural constraints of feature embeddings through a loss function constructed via a combinatorial graph-matching objective; and (iii) it can be trained efficiently end-to-end using modern gradient-estimation techniques for black-box solvers. We thoroughly evaluate the proposed LVM-Med on 15 downstream medical tasks ranging from segmentation and classification to object detection, and both for the in and out-of-distribution settings. LVM-Med empirically outperforms a number of state-of-the-art supervised, self-supervised, and foundation models. For challenging tasks such as Brain Tumor Classification or Diabetic Retinopathy Grading, LVM-Med improves previous vision-language models trained on 1 billion masks by 6-7% while using only a ResNet-50.
Authors: Yujun Shi, Chuhui Xue, Jiachun Pan, Wenqing Zhang, Vincent Y. F. Tan, Song Bai
Precise and controllable image editing is a challenging task that has attracted significant attention. Recently, DragGAN enables an interactive point-based image editing framework and achieves impressive editing results with pixel-level precision. However, since this method is based on generative adversarial networks (GAN), its generality is upper-bounded by the capacity of the pre-trained GAN models. In this work, we extend such an editing framework to diffusion models and propose DragDiffusion. By leveraging large-scale pretrained diffusion models, we greatly improve the applicability of interactive point-based editing in real world scenarios. While most existing diffusion-based image editing methods work on text embeddings, DragDiffusion optimizes the diffusion latent to achieve precise spatial control. Although diffusion models generate images in an iterative manner, we empirically show that optimizing diffusion latent at one single step suffices to generate coherent results, enabling DragDiffusion to complete high-quality editing efficiently. Extensive experiments across a wide range of challenging cases (e.g., multi-objects, diverse object categories, various styles, etc.) demonstrate the versatility and generality of DragDiffusion. Code: https://github.com/Yujun-Shi/DragDiffusion.
Authors: Peng Sun, Yongbin Zheng, Wenqi Wu, Wanying Xu, Shengjian Bai
Arbitrary-oriented object detection is a relatively emerging but challenging task. Although remarkable progress has been made, there still remain many unsolved issues due to the large diversity of patterns in orientation, scale, aspect ratio, and visual appearance of objects in aerial images. Most of the existing methods adopt a coarse-grained fixed label assignment strategy and suffer from the inconsistency between the classification score and localization accuracy. First, to align the metric inconsistency between sample selection and regression loss calculation caused by fixed IoU strategy, we introduce affine transformation to evaluate the quality of samples and propose a distance-based label assignment strategy. The proposed metric-aligned selection (MAS) strategy can dynamically select samples according to the shape and rotation characteristic of objects. Second, to further address the inconsistency between classification and localization, we propose a critical feature sampling (CFS) module, which performs localization refinement on the sampling location for classification task to extract critical features accurately. Third, we present a scale-controlled smooth $L_1$ loss (SC-Loss) to adaptively select high quality samples by changing the form of regression loss function based on the statistics of proposals during training. Extensive experiments are conducted on four challenging rotated object detection datasets DOTA, FAIR1M-1.0, HRSC2016, and UCAS-AOD. The results show the state-of-the-art accuracy of the proposed detector.
Authors: Balamurali Murugesan, Rukhshanda Hussain, Rajarshi Bhattacharya, Ismail Ben Ayed, Jose Dolz
Recently, CLIP-based approaches have exhibited remarkable performance on generalization and few-shot learning tasks, fueled by the power of contrastive language-vision pre-training. In particular, prompt tuning has emerged as an effective strategy to adapt the pre-trained language-vision models to downstream tasks by employing task-related textual tokens. Motivated by this progress, in this work we question whether other fundamental problems, such as weakly supervised semantic segmentation (WSSS), can benefit from prompt tuning. Our findings reveal two interesting observations that shed light on the impact of prompt tuning on WSSS. First, modifying only the class token of the text prompt results in a greater impact on the Class Activation Map (CAM), compared to arguably more complex strategies that optimize the context. And second, the class token associated with the image ground truth does not necessarily correspond to the category that yields the best CAM. Motivated by these observations, we introduce a novel approach based on a PrOmpt cLass lEarning (POLE) strategy. Through extensive experiments we demonstrate that our simple, yet efficient approach achieves SOTA performance in a well-known WSSS benchmark. These results highlight not only the benefits of language-vision models in WSSS but also the potential of prompt learning for this problem. The code is available at https://github.com/rB080/WSS_POLE.
Authors: Minh-Quan Dao, Julie Stephany Berrio, Vincent Frémont, Mao Shan, Elwan Héry, Stewart Worrall
Occlusion is a major challenge for LiDAR-based object detection methods. This challenge becomes safety-critical in urban traffic where the ego vehicle must have reliable object detection to avoid collision while its field of view is severely reduced due to the obstruction posed by a large number of road users. Collaborative perception via Vehicle-to-Everything (V2X) communication, which leverages the diverse perspective thanks to the presence at multiple locations of connected agents to form a complete scene representation, is an appealing solution. State-of-the-art V2X methods resolve the performance-bandwidth tradeoff using a mid-collaboration approach where the Bird-Eye View images of point clouds are exchanged so that the bandwidth consumption is lower than communicating point clouds as in early collaboration, and the detection performance is higher than late collaboration, which fuses agents' output, thanks to a deeper interaction among connected agents. While achieving strong performance, the real-world deployment of most mid-collaboration approaches is hindered by their overly complicated architectures, involving learnable collaboration graphs and autoencoder-based compressor/ decompressor, and unrealistic assumptions about inter-agent synchronization. In this work, we devise a simple yet effective collaboration method that achieves a better bandwidth-performance tradeoff than prior state-of-the-art methods while minimizing changes made to the single-vehicle detection models and relaxing unrealistic assumptions on inter-agent synchronization. Experiments on the V2X-Sim dataset show that our collaboration method achieves 98\% of the performance of an early-collaboration method, while only consuming the equivalent bandwidth of a late-collaboration method.
Authors: Yunqing Zhao, Keshigeyan Chandrasegaran, Abdollahzadeh Milad, Chao Du, Tianyu Pang, Ruoteng Li, Henghui Ding, Ngai-Man Cheung
Few-shot image generation (FSIG) aims to learn to generate new and diverse images given few (e.g., 10) training samples. Recent work has addressed FSIG by leveraging a GAN pre-trained on a large-scale source domain and adapting it to the target domain with few target samples. Central to recent FSIG methods are knowledge preservation criteria, which select and preserve a subset of source knowledge to the adapted model. However, a major limitation of existing methods is that their knowledge preserving criteria consider only source domain/task and fail to consider target domain/adaptation in selecting source knowledge, casting doubt on their suitability for setups of different proximity between source and target domain. Our work makes two contributions. Firstly, we revisit recent FSIG works and their experiments. We reveal that under setups which assumption of close proximity between source and target domains is relaxed, many existing state-of-the-art (SOTA) methods which consider only source domain in knowledge preserving perform no better than a baseline method. As our second contribution, we propose Adaptation-Aware kernel Modulation (AdAM) for general FSIG of different source-target domain proximity. Extensive experiments show that AdAM consistently achieves SOTA performance in FSIG, including challenging setups where source and target domains are more apart.
Authors: Raha Ahmadi, Mohammad Javad Rajabi, Mohammad Khalooie, Mohammad Sabokrou
Deep learning models have demonstrated remarkable capabilities in learning complex patterns and concepts from training data. However, recent findings indicate that these models tend to rely heavily on simple and easily discernible features present in the background of images rather than the main concepts or objects they are intended to classify. This phenomenon poses a challenge to image classifiers as the crucial elements of interest in images may be overshadowed. In this paper, we propose a novel approach to address this issue and improve the learning of main concepts by image classifiers. Our central idea revolves around concurrently guiding the model's attention toward the foreground during the classification task. By emphasizing the foreground, which encapsulates the primary objects of interest, we aim to shift the focus of the model away from the dominant influence of the background. To accomplish this, we introduce a mechanism that encourages the model to allocate sufficient attention to the foreground. We investigate various strategies, including modifying the loss function or incorporating additional architectural components, to enable the classifier to effectively capture the primary concept within an image. Additionally, we explore the impact of different foreground attention mechanisms on model performance and provide insights into their effectiveness. Through extensive experimentation on benchmark datasets, we demonstrate the efficacy of our proposed approach in improving the classification accuracy of image classifiers. Our findings highlight the importance of foreground attention in enhancing model understanding and representation of the main concepts within images. The results of this study contribute to advancing the field of image classification and provide valuable insights for developing more robust and accurate deep-learning models.
Authors: Zhijie Rao, Jingcai Guo, Luyao Tang, Yue Huang, Xinghao Ding, Song Guo
This paper provides a novel framework for single-domain generalized object detection (i.e., Single-DGOD), where we are interested in learning and maintaining the semantic structures of self-augmented compound cross-domain samples to enhance the model's generalization ability. Different from DGOD trained on multiple source domains, Single-DGOD is far more challenging to generalize well to multiple target domains with only one single source domain. Existing methods mostly adopt a similar treatment from DGOD to learn domain-invariant features by decoupling or compressing the semantic space. However, there may have two potential limitations: 1) pseudo attribute-label correlation, due to extremely scarce single-domain data; and 2) the semantic structural information is usually ignored, i.e., we found the affinities of instance-level semantic relations in samples are crucial to model generalization. In this paper, we introduce Semantic Reasoning with Compound Domains (SRCD) for Single-DGOD. Specifically, our SRCD contains two main components, namely, the texture-based self-augmentation (TBSA) module, and the local-global semantic reasoning (LGSR) module. TBSA aims to eliminate the effects of irrelevant attributes associated with labels, such as light, shadow, color, etc., at the image level by a light-yet-efficient self-augmentation. Moreover, LGSR is used to further model the semantic relationships on instance features to uncover and maintain the intrinsic semantic structures. Extensive experiments on multiple benchmarks demonstrate the effectiveness of the proposed SRCD.
Authors: Duong Q. Nguyen, Thinh D. Le, Phuong D. Nguyen, Nga T.K. Le, H. Nguyen-Xuan
Facial wound segmentation plays a crucial role in preoperative planning and optimizing patient outcomes in various medical applications. In this paper, we propose an efficient approach for automating 3D facial wound segmentation using a two-stream graph convolutional network. Our method leverages the Cir3D-FaIR dataset and addresses the challenge of data imbalance through extensive experimentation with different loss functions. To achieve accurate segmentation, we conducted thorough experiments and selected a high-performing model from the trained models. The selected model demonstrates exceptional segmentation performance for complex 3D facial wounds. Furthermore, based on the segmentation model, we propose an improved approach for extracting 3D facial wound fillers and compare it to the results of the previous study. Our method achieved a remarkable accuracy of 0.9999986\% on the test suite, surpassing the performance of the previous method. From this result, we use 3D printing technology to illustrate the shape of the wound filling. The outcomes of this study have significant implications for physicians involved in preoperative planning and intervention design. By automating facial wound segmentation and improving the accuracy of wound-filling extraction, our approach can assist in carefully assessing and optimizing interventions, leading to enhanced patient outcomes. Additionally, it contributes to advancing facial reconstruction techniques by utilizing machine learning and 3D bioprinting for printing skin tissue implants. Our source code is available at \url{https://github.com/SIMOGroup/WoundFilling3D}.
Authors: Jiahao Li, Yuanyou Xu, Zongxin Yang, Yi Yang, Yueting Zhuang
The Associating Objects with Transformers (AOT) framework has exhibited exceptional performance in a wide range of complex scenarios for video object segmentation. In this study, we introduce MSDeAOT, a variant of the AOT series that incorporates transformers at multiple feature scales. Leveraging the hierarchical Gated Propagation Module (GPM), MSDeAOT efficiently propagates object masks from previous frames to the current frame using a feature scale with a stride of 16. Additionally, we employ GPM in a more refined feature scale with a stride of 8, leading to improved accuracy in detecting and tracking small objects. Through the implementation of test-time augmentations and model ensemble techniques, we achieve the top-ranking position in the EPIC-KITCHEN VISOR Semi-supervised Video Object Segmentation Challenge.
Authors: Fatemeh Farokhmanesh, Kevin Höhlein, Christoph Neuhauser, Rüdiger Westermann
We present the first neural network that has learned to compactly represent and can efficiently reconstruct the statistical dependencies between the values of physical variables at different spatial locations in large 3D simulation ensembles. Going beyond linear dependencies, we consider mutual information as a measure of non-linear dependence. We demonstrate learning and reconstruction with a large weather forecast ensemble comprising 1000 members, each storing multiple physical variables at a 250 x 352 x 20 simulation grid. By circumventing compute-intensive statistical estimators at runtime, we demonstrate significantly reduced memory and computation requirements for reconstructing the major dependence structures. This enables embedding the estimator into a GPU-accelerated direct volume renderer and interactively visualizing all mutual dependencies for a selected domain point.
Authors: Lukas Muttenthaler, Robert A. Vandermeulen, Qiuyi Zhang, Thomas Unterthiner, Klaus-Robert Müller
Model overconfidence and poor calibration are common in machine learning and difficult to account for when applying standard empirical risk minimization. In this work, we propose a novel method to alleviate these problems that we call odd-$k$-out learning (OKO), which minimizes the cross-entropy error for sets rather than for single examples. This naturally allows the model to capture correlations across data examples and achieves both better accuracy and calibration, especially in limited training data and class-imbalanced regimes. Perhaps surprisingly, OKO often yields better calibration even when training with hard labels and dropping any additional calibration parameter tuning, such as temperature scaling. We provide theoretical justification, establishing that OKO naturally yields better calibration, and provide extensive experimental analyses that corroborate our theoretical findings. We emphasize that OKO is a general framework that can be easily adapted to many settings and the trained model can be applied to single examples at inference time, without introducing significant run-time overhead or architecture changes.
Authors: Jiamiao Zhang, Yichen Chi, Jun Lyu, Wenming Yang, Yapeng Tian
Limited by imaging systems, the reconstruction of Magnetic Resonance Imaging (MRI) images from partial measurement is essential to medical imaging research. Benefiting from the diverse and complementary information of multi-contrast MR images in different imaging modalities, multi-contrast Super-Resolution (SR) reconstruction is promising to yield SR images with higher quality. In the medical scenario, to fully visualize the lesion, radiologists are accustomed to zooming the MR images at arbitrary scales rather than using a fixed scale, as used by most MRI SR methods. In addition, existing multi-contrast MRI SR methods often require a fixed resolution for the reference image, which makes acquiring reference images difficult and imposes limitations on arbitrary scale SR tasks. To address these issues, we proposed an implicit neural representations based dual-arbitrary multi-contrast MRI super-resolution method, called Dual-ArbNet. First, we decouple the resolution of the target and reference images by a feature encoder, enabling the network to input target and reference images at arbitrary scales. Then, an implicit fusion decoder fuses the multi-contrast features and uses an Implicit Decoding Function~(IDF) to obtain the final MRI SR results. Furthermore, we introduce a curriculum learning strategy to train our network, which improves the generalization and performance of our Dual-ArbNet. Extensive experiments in two public MRI datasets demonstrate that our method outperforms state-of-the-art approaches under different scale factors and has great potential in clinical practice.
Authors: Yuanyou Xu, Jiahao Li, Zongxin Yang, Yi Yang, Yueting Zhuang
The Associating Objects with Transformers (AOT) framework has exhibited exceptional performance in a wide range of complex scenarios for video object tracking and segmentation. In this study, we convert the bounding boxes to masks in reference frames with the help of the Segment Anything Model (SAM) and Alpha-Refine, and then propagate the masks to the current frame, transforming the task from Video Object Tracking (VOT) to video object segmentation (VOS). Furthermore, we introduce MSDeAOT, a variant of the AOT series that incorporates transformers at multiple feature scales. MSDeAOT efficiently propagates object masks from previous frames to the current frame using two feature scales of 16 and 8. As a testament to the effectiveness of our design, we achieved the 1st place in the EPIC-KITCHENS TREK-150 Object Tracking Challenge.
Authors: Harbir Antil, David Sayre
This paper explores the application of event-based cameras in the domains of image segmentation and motion estimation. These cameras offer a groundbreaking technology by capturing visual information as a continuous stream of asynchronous events, departing from the conventional frame-based image acquisition. We introduce a Generalized Nash Equilibrium based framework that leverages the temporal and spatial information derived from the event stream to carry out segmentation and velocity estimation. To establish the theoretical foundations, we derive an existence criteria and propose a multi-level optimization method for calculating equilibrium. The efficacy of this approach is shown through a series of experiments.
Authors: Peter Tu, Zhaoyuan Yang, Richard Hartley, Zhiwei Xu, Jing Zhang, Dylan Campbell, Jaskirat Singh, Tianyu Wang
This paper begins with a description of methods for estimating probability density functions for images that reflects the observation that such data is usually constrained to lie in restricted regions of the high-dimensional image space - not every pattern of pixels is an image. It is common to say that images lie on a lower-dimensional manifold in the high-dimensional space. However, although images may lie on such lower-dimensional manifolds, it is not the case that all points on the manifold have an equal probability of being images. Images are unevenly distributed on the manifold, and our task is to devise ways to model this distribution as a probability distribution. In pursuing this goal, we consider generative models that are popular in AI and computer vision community. For our purposes, generative/probabilistic models should have the properties of 1) sample generation: it should be possible to sample from this distribution according to the modelled density function, and 2) probability computation: given a previously unseen sample from the dataset of interest, one should be able to compute the probability of the sample, at least up to a normalising constant. To this end, we investigate the use of methods such as normalising flow and diffusion models. We then show that such probabilistic descriptions can be used to construct defences against adversarial attacks. In addition to describing the manifold in terms of density, we also consider how semantic interpretations can be used to describe points on the manifold. To this end, we consider an emergent language framework which makes use of variational encoders to produce a disentangled representation of points that reside on a given manifold. Trajectories between points on a manifold can then be described in terms of evolving semantic descriptions.
Authors: Yijie Deng, Lei Han, Tianpeng Lin, Lin Li, Jinzhi Zhang, Lu Fang
With the rise of Extended Reality (XR) technology, there is a growing need for real-time light field generation from sparse view inputs. Existing methods can be classified into offline techniques, which can generate high-quality novel views but at the cost of long inference/training time, and online methods, which either lack generalizability or produce unsatisfactory results. However, we have observed that the intrinsic sparse manifold of Multi-plane Images (MPI) enables a significant acceleration of light field generation while maintaining rendering quality. Based on this insight, we introduce EffLiFe, a novel light field optimization method, which leverages the proposed Hierarchical Sparse Gradient Descent (HSGD) to produce high-quality light fields from sparse view images in real time. Technically, the coarse MPI of a scene is first generated using a 3D CNN, and it is further sparsely optimized by focusing only on important MPI gradients in a few iterations. Nevertheless, relying solely on optimization can lead to artifacts at occlusion boundaries. Therefore, we propose an occlusion-aware iterative refinement module that removes visual artifacts in occluded regions by iteratively filtering the input. Extensive experiments demonstrate that our method achieves comparable visual quality while being 100x faster on average than state-of-the-art offline methods and delivering better performance (about 2 dB higher in PSNR) compared to other online approaches.
Authors: Ludovica Schaerf, Carina Popovici, Eric Postma
In recent years, Transformers, initially developed for language, have been successfully applied to visual tasks. Vision Transformers have been shown to push the state-of-the-art in a wide range of tasks, including image classification, object detection, and semantic segmentation. While ample research has shown promising results in art attribution and art authentication tasks using Convolutional Neural Networks, this paper examines if the superiority of Vision Transformers extends to art authentication, improving, thus, the reliability of computer-based authentication of artworks. Using a carefully compiled dataset of authentic paintings by Vincent van Gogh and two contrast datasets, we compare the art authentication performances of Swin Transformers with those of EfficientNet. Using a standard contrast set containing imitations and proxies (works by painters with styles closely related to van Gogh), we find that EfficientNet achieves the best performance overall. With a contrast set that only consists of imitations, we find the Swin Transformer to be superior to EfficientNet by achieving an authentication accuracy of over 85%. These results lead us to conclude that Vision Transformers represent a strong and promising contender in art authentication, particularly in enhancing the computer-based ability to detect artistic imitations.
Authors: Jishnu Jaykumar P, Kamalesh Palanisamy, Yu-Wei Chao, Xinya Du, Yu Xiang
We propose a novel framework for few-shot learning by leveraging large-scale vision-language models such as CLIP. Motivated by the unimodal prototypical networks for few-shot learning, we introduce PROTO-CLIP that utilizes image prototypes and text prototypes for few-shot learning. Specifically, PROTO-CLIP adapts the image encoder and text encoder in CLIP in a joint fashion using few-shot examples. The two encoders are used to compute prototypes of image classes for classification. During adaptation, we propose aligning the image and text prototypes of corresponding classes. Such a proposed alignment is beneficial for few-shot classification due to the contributions from both types of prototypes. We demonstrate the effectiveness of our method by conducting experiments on benchmark datasets for few-shot learning as well as in the real world for robot perception.