Total: 146
Tracking individuals is a vital part of many experiments conducted to understand collective behaviour. Ants are the paradigmatic model system for such experiments but their lack of individually distinguishing visual features and their high colony densities make it extremely difficult to perform reliable racking automatically. Additionally, the wide diversity of their species' appearances makes a generalized approach even harder. In this paper, we propose a data-driven multi-object tracker that, for the first time, employs domain adaptation to achieve the required generalisation. This approach is built upon a joint-detection-and-tracking framework that is extended by a set of domain discriminator modules integrating an adversarial training strategy in addition to the tracking loss. In addition to this novel domain-adaptive tracking framework, we present a new dataset and a benchmark for the ant tracking problem. The dataset contains 57 video sequences with full trajectory annotation, including 30k frames captured from two different ant species moving on different background patterns. It comprises 33 and 24 sequences for source and target domains, respectively. We compare our proposed framework against other domain-adaptive and non-domain-adaptive multi-object tracking baselines using this dataset and show that incorporating domain adaptation at multiple levels of the tracking pipeline yields significant improvements. The code and the dataset are available at https://github.com/chamathabeysinghe/da-tracker.
Text-based person search aims to retrieve the specified person images given a textual description. The key to tackling such a challenging task is to learn powerful multi-modal representations. Towards this, we propose a Relation and Sensitivity aware representation learning method (RaSa), including two novel tasks: Relation-Aware learning (RA) and Sensitivity-Aware learning (SA). For one thing, existing methods cluster representations of all positive pairs without distinction and overlook the noise problem caused by the weak positive pairs where the text and the paired image have noise correspondences, thus leading to overfitting learning. RA offsets the overfitting risk by introducing a novel positive relation detection task (i.e., learning to distinguish strong and weak positive pairs). For another thing, learning invariant representation under data augmentation (i.e., being insensitive to some transformations) is a general practice for improving representation's robustness in existing methods. Beyond that, we encourage the representation to perceive the sensitive transformation by SA (i.e., learning to detect the replaced words), thus promoting the representation's robustness. Experiments demonstrate that RaSa outperforms existing state-of-the-art methods by 6.94%, 4.45% and 15.35% in terms of Rank@1 on CUHK-PEDES, ICFG-PEDES and RSTPReid datasets, respectively. Code is available at: https://github.com/Flame-Chasers/RaSa.
Deep convolutional neural networks (CNNs) have achieved unprecedented success in single image super-resolution over the past few years. Meanwhile, there is an increasing demand for single image super-resolution with arbitrary scale factors in real-world scenarios. Many approaches adopt scale-specific multi-path learning to cope with multi-scale super-resolution with a single network. However, these methods require a large number of parameters. To achieve a better balance between the reconstruction quality and parameter amounts, we proposes a learnable interpolation method that leverages the advantages of neural networks and interpolation methods to tackle the scale-arbitrary super-resolution task. The scale factor is treated as a function parameter for generating the kernel weights for the learnable interpolation. We demonstrate that the learnable interpolation builds a bridge between neural networks and traditional interpolation methods. Experiments show that the proposed learnable interpolation requires much fewer parameters and outperforms state-of-the-art super-resolution methods.
Pansharpening is to fuse a panchromatic (PAN) image with a multispectral (MS) image to obtain a high-spatial-resolution multispectral (HRMS) image. The deep learning-based pansharpening methods usually apply the convolution operation to extract features and only consider the similarity of gradient information between PAN and HRMS images, resulting in the problems of edge blur and spectral distortion in the fusion results. To solve this problem, a multi-supervised mask protection network (MMPN) is proposed to prevent spatial information from being damaged and overcome spectral distortion in the learning process. Firstly, by analyzing the relationships between high-resolution images and corresponding degraded images, a mask protection strategy (MPS) for edge protection is designed to guide the recovery of fused images. Then, based on the MPS, an MMPN containing four branches is constructed to generate the fusion and mask protection images. In MMPN, each branch employs a dual-stream multi-scale feature fusion module (DMFFM), which is built to extract and fuse the features of two input images. Finally, different loss terms are defined for the four branches, and combined into a joint loss function to realize network training. Experiments on simulated and real satellite datasets show that our method is superior to state-of-the-art methods both subjectively and objectively.
Human pose estimation is a challenging task due to its structured data sequence nature. Existing methods primarily focus on pair-wise interaction of body joints, which is insufficient for scenarios involving overlapping joints and rapidly changing poses. To overcome these issues, we introduce a novel approach, the High-order Directed Transformer (HDFormer), which leverages high-order bone and joint relationships for improved pose estimation. Specifically, HDFormer incorporates both self-attention and high-order attention to formulate a multi-order attention module. This module facilitates first-order "joint-joint", second-order "bone-joint", and high-order "hyperbone-joint" interactions, effectively addressing issues in complex and occlusion-heavy situations. In addition, modern CNN techniques are integrated into the transformer-based architecture, balancing the trade-off between performance and efficiency. HDFormer significantly outperforms state-of-the-art (SOTA) models on Human3.6M and MPI-INF-3DHP datasets, requiring only 1/10 of the parameters and significantly lower computational costs. Moreover, HDFormer demonstrates broad real-world applicability, enabling real-time, accurate 3D pose estimation. The source code is in https://github.com/hyer/HDFormer.
Most infrared small target detection (ISTD) networks focus on building effective neural blocks or feature fusion modules but none describes the ISTD process from the image evolution perspective. The directional evolution of image pixels influenced by convolution, pooling and surrounding pixels is analogous to the movement of fluid elements constrained by surrounding variables ang particles. Inspired by this, we explore a novel research routine by abstracting the movement of pixels in the ISTD process as the flow of fluid in fluid dynamics (FD). Specifically, a new Fluid Dynamics-Inspired Network (FDI-Net) is devised for ISTD. Based on Taylor Central Difference (TCD) method, the TCD feature extraction block is designed, where convolution and Transformer structures are combined for local and global information. The pixel motion equation during the ISTD process is derived from the Navier–Stokes (N-S) equation, constructing a N-S Refinement Module that refines extracted features with edge details. Thus, the TCD feature extraction block determines the primary movement direction of pixels during detection, while the N-S Refinement Module corrects some skewed directions of the pixel stream to supplement the edge details. Experiments on IRSTD-1k and SIRST demonstrate that our method achieves SOTA performance in terms of evaluation metrics.
The core of Multi-view Stereo(MVS) is the matching process among reference and source pixels. Cost aggregation plays a significant role in this process, while previous methods focus on handling it via CNNs. This may inherit the natural limitation of CNNs that fail to discriminate repetitive or incorrect matches due to limited local receptive fields. To handle the issue, we aim to involve Transformer into cost aggregation. However, another problem may occur due to the quadratically growing computational complexity caused by Transformer, resulting in memory overflow and inference latency. In this paper, we overcome these limits with an efficient Transformer-based cost aggregation network, namely CostFormer. The Residual Depth-Aware Cost Transformer(RDACT) is proposed to aggregate long-range features on cost volume via self-attention mechanisms along the depth and spatial dimensions. Furthermore, Residual Regression Transformer(RRT) is proposed to enhance spatial attention. The proposed method is a universal plug-in to improve learning-based MVS methods.
The performance of existing supervised neuron segmentation methods is highly dependent on the number of accurate annotations, especially when applied to large scale electron microscopy (EM) data. By extracting semantic information from unlabeled data, self-supervised methods can improve the performance of downstream tasks, among which the mask image model (MIM) has been widely used due to its simplicity and effectiveness in recovering original information from masked images. However, due to the high degree of structural locality in EM images, as well as the existence of considerable noise, many voxels contain little discriminative information, making MIM pretraining inefficient on the neuron segmentation task. To overcome this challenge, we propose a decision-based MIM that utilizes reinforcement learning (RL) to automatically search for optimal image masking ratio and masking strategy. Due to the vast exploration space, using single-agent RL for voxel prediction is impractical. Therefore, we treat each input patch as an agent with a shared behavior policy, allowing for multi-agent collaboration. Furthermore, this multi-agent model can capture dependencies between voxels, which is beneficial for the downstream segmentation task. Experiments conducted on representative EM datasets demonstrate that our approach has a significant advantage over alternative self-supervised methods on the task of neuron segmentation. Code is available at https://github.com/ydchen0806/dbMiM.
Point cloud completion aims at estimating the complete data of objects from degraded observations. Despite existing completion methods achieving impressive performances, they rely heavily on degraded-complete data pairs for supervision. In this work, we propose a novel framework named Null-Space Diffusion Sampling (NSDS) to solve the point cloud completion task in a zero-shot manner. By leveraging a pre-trained point cloud diffusion model as the off-the-shelf generator, our sampling approach can generate desired completion outputs with the guidance of the observed degraded data without any extra training. Furthermore, we propose a tolerant loop mechanism to improve the quality of completion results for hard cases. Experimental results demonstrate our zero-shot framework achieves superior completion performance than unsupervised methods and comparable performance to supervised methods in various degraded situations.
Image ordinal regression has been mainly studied along the line of exploiting the order of categories. However, the issues of class imbalance and category overlap that are very common in ordinal regression were largely overlooked. As a result, the performance on minority categories is often unsatisfactory. In this paper, we propose a novel framework called CIG based on controllable image generation to directly tackle these two issues. Our main idea is to generate extra training samples with specific labels near category boundaries, and the sample generation is biased toward the less-represented categories. To achieve controllable image generation, we seek to separate structural and categorical information of images based on structural similarity, categorical similarity, and reconstruction constraints. We evaluate the effectiveness of our new CIG approach in three different image ordinal regression scenarios. The results demonstrate that CIG can be flexibly integrated with off-the-shelf image encoders or ordinal regression models to achieve improvement, and further, the improvement is more significant for minority categories.
The top-down and bottom-up methods are two mainstreams of referring segmentation, while both methods have their own intrinsic weaknesses. Top-down methods are chiefly disturbed by Polar Negative (PN) errors owing to the lack of fine-grained cross-modal alignment. Bottom-up methods are mainly perturbed by Inferior Positive (IP) errors due to the lack of prior object information. Nevertheless, we discover that two types of methods are highly complementary for restraining respective weaknesses but the direct average combination leads to harmful interference. In this context, we build Win-win Cooperation (WiCo) to exploit complementary nature of two types of methods on both interaction and integration aspects for achieving a win-win improvement. For the interaction aspect, Complementary Feature Interaction (CFI) introduces prior object information to bottom-up branch and provides fine-grained information to top-down branch for complementary feature enhancement. For the integration aspect, Gaussian Scoring Integration (GSI) models the gaussian performance distributions of two branches and weighted integrates results by sampling confident scores from the distributions. With our WiCo, several prominent bottom-up and top-down combinations achieve remarkable improvements on three common datasets with reasonable extra costs, which justifies effectiveness and generality of our method.
As a long-standing task, image restoration aims to recover the latent sharp image from its degraded counterpart. In recent years, owing to the strong ability of self-attention in capturing long-range dependencies, Transformer based methods have achieved promising performance on multifarious image restoration tasks. However, the canonical self-attention leads to quadratic complexity with respect to input size, hindering its further applications in image restoration. In this paper, we propose a Strip Attention Network (SANet) for image restoration to integrate information in a more efficient and effective manner. Specifically, a strip attention unit is proposed to harvest the contextual information for each pixel from its adjacent pixels in the same row or column. By employing this operation in different directions, each location can perceive information from an expanded region. Furthermore, we apply various receptive fields in different feature groups to enhance representation learning. Incorporating these designs into a U-shaped backbone, our SANet performs favorably against state-of-the-art algorithms on several image restoration tasks. The code is available at https://github.com/c-yn/SANet.
The long-tail effect is a common issue that limits the performance of deep learning models on real-world datasets. Character image datasets are also affected by such unbalanced data distribution due to differences in character usage frequency. Thus, current character recognition methods are limited when applied in the real world, especially for the categories in the tail that lack training samples, e.g., uncommon characters. In this paper, we propose a zero-shot character recognition framework via radical-based reasoning, called RZCR, to improve the recognition performance of few-sample character categories in the tail. Specifically, we exploit radicals, the graphical units of characters, by decomposing and reconstructing characters according to orthography. RZCR consists of a visual semantic fusion-based radical information extractor (RIE) and a knowledge graph character reasoner (KGR). RIE aims to recognize candidate radicals and their possible structural relations from character images in parallel. The results are then fed into KGR to recognize the target character by reasoning with a knowledge graph. We validate our method on multiple datasets, and RZCR shows promising experimental results, especially on few-sample character datasets.
Semi-supervised semantic segmentation methods are the main solution to alleviate the problem of high annotation consumption in semantic segmentation. However, the class imbalance problem makes the model favor the head classes with sufficient training samples, resulting in poor performance of the tail classes. To address this issue, we propose a Decoupled Semi-Supervise Semantic Segmentation (DeS4) framework based on the teacher-student model. Specifically, we first propose a decoupling training strategy to split the training of the encoder and segmentation decoder, aiming at a balanced decoder. Then, a non-learnable prototype-based segmentation head is proposed to regularize the category representation distribution consistency and perform a better connection between the teacher model and the student model. Furthermore, a Multi-Entropy Sampling (MES) strategy is proposed to collect pixel representation for updating the shared prototype to get a class-unbiased head. We conduct extensive experiments of the proposed DeS4 on two challenging benchmarks (PASCAL VOC 2012 and Cityscapes) and achieve remarkable improvements over the previous state-of-the-art methods.
The performance of nighttime semantic segmentation has been significantly improved thanks to recent unsupervised methods. However, these methods still suffer from complex domain gaps, i.e., the challenging illumination gap and the inherent dataset gap. In this paper, we propose the illumination-coupled domain adaptation framework(ICDA) to effectively avoid the illumination gap and mitigate the dataset gap by coupling daytime and nighttime images as a whole with semantic relevance. Specifically, we first design a new composite enhancement method(CEM) that considers not only illumination but also spatial consistency to construct the source and target domain pairs, which provides the basic adaptation unit for our ICDA. Next, to avoid the illumination gap, we devise the Deformable Attention Relevance(DAR) module to capture the semantic relevance inside each domain pair, which can couple the daytime and nighttime images at the feature level and adaptively guide the predictions of nighttime images. Besides, to mitigate the dataset gap and acquire domain-invariant semantic relevance, we propose the Prototype-based Class Alignment(PCA) module, which improves the usage of category information and performs fine-grained alignment. Extensive experiments show that our method reduces the complex domain gaps and achieves state-of-the-art performance for nighttime semantic segmentation. Our code is available at https://github.com/chenghaoDong666/ICDA.
Recently, techniques utilizing frequency-based methods have gained significant attention, as they exhibit exceptional restoration capabilities for detail and structure in video super-resolution tasks. However, most of these frequency-based methods mainly have three major limitations: 1) insufficient exploration of object motion information, 2) inadequate enhancement for high-fidelity regions, and 3) loss of spatial information during convolution. In this paper, we propose a novel network, Directional Frequency Video Super-Resolution (DFVSR), to address these limitations. Specifically, we reconsider object motion from a new perspective and propose Directional Frequency Representation (DFR), which not only borrows the property of frequency representation of detail and structure information but also contains the direction information of the object motion that is extremely significant in videos. Based on this representation, we propose a Directional Frequency-Enhanced Alignment (DFEA) to use double enhancements of task-related information for ensuring the retention of high-fidelity frequency regions to generate the high-quality alignment feature. Furthermore, we design a novel Asymmetrical U-shaped network architecture to progressively fuse these alignment features and output the final output. This architecture enables the intercommunication of the same level of resolution in the encoder and decoder to achieve the supplement of spatial information. Powered by the above designs, our method achieves superior performance over state-of-the-art models on both quantitative and qualitative evaluations.
Video action segmentation under timestamp supervision has recently received much attention due to lower annotation costs. Most existing methods generate pseudo-labels for all frames in each video to train the segmentation model. However, these methods suffer from incorrect pseudo-labels, especially for the semantically unclear frames in the transition region between two consecutive actions, which we call ambiguous intervals. To address this issue, we propose a novel framework from the perspective of clustering, which includes the following two parts. First, pseudo-label ensembling generates incomplete but high-quality pseudo-label sequences, where the frames in ambiguous intervals have no pseudo-labels. Second, iterative clustering iteratively propagates the pseudo-labels to the ambiguous intervals by clustering, and thus updates the pseudo-label sequences to train the model. We further introduce a clustering loss, which encourages the features of frames within the same action segment more compact. Extensive experiments show the effectiveness of our method.
Semi-supervised deep facial expression recognition (SS-DFER) has recently attracted rising research interest due to its more practical setting of abundant unlabeled data. However, there are two main problems unconsidered in current SS-DFER methods: 1) label ambiguity, i.e., given labels mismatch with facial expressions; 2) inefficient utilization of unlabeled data with low-confidence. In this paper, we propose a novel SS-DFER method, including a Label DIsambiguation module and a PrOgressive Negative Learning module, namely LION, to simultaneously address both problems. Specifically, the label disambiguation module operates on labeled data, including data with accurate labels (clear data) and ambiguous labels (ambiguous data). It first uses clear data to calculate prototypes for all the expression classes, and then re-assign a candidate label set to all the ambiguous data. Based on the prototypes and the candidate label set, the ambiguous data can be relabeled more accurately. As for unlabeled data with low-confidence, the progressive negative learning module is developed to iteratively mine more complete complementary labels, which can guide the model to reduce the association between data and corresponding complementary labels. Experiments on three challenging datasets show that our method significantly outperforms the current state-of-the-art approaches in SS-DFER and surpasses fully-supervised baselines. Code will be available at https://github.com/NUM-7/LION.
Recent works reveal that adversarial augmentation benefits the generalization of neural networks (NNs) if used in an appropriate manner. In this paper, we introduce Temporal Adversarial Augmentation (TA), a novel video augmentation technique that utilizes temporal attention. Unlike conventional adversarial augmentation, TA is specifically designed to shift the attention distributions of neural networks with respect to video clips by maximizing a temporal-related loss function. We demonstrate that TA will obtain diverse temporal views, which significantly affect the focus of neural networks. Training with these examples remedies the flaw of unbalanced temporal information perception and enhances the ability to defend against temporal shifts, ultimately leading to better generalization. To leverage TA, we propose Temporal Video Adversarial Fine-tuning (TAF) framework for improving video representations. TAF is a model-agnostic, generic, and interpretability-friendly training strategy. We evaluate TAF with four powerful models (TSM, GST, TAM, and TPN) over three challenging temporal-related benchmarks (Something-something V1&V2 and diving48). Experimental results demonstrate that TAF effectively improves the test accuracy of these models with notable margins without introducing additional parameters or computational costs. As a byproduct, TAF also improves the robustness under out-of-distribution (OOD) settings. Code is available at https://github.com/jinhaoduan/TAF.
Glass-like objects are widespread in daily life but remain intractable to be segmented for most existing methods. The transparent property makes it difficult to be distinguished from background, while the tiny separation boundary further impedes the acquisition of their exact contour. In this paper, by revealing the key co-evolution demand of semantic and boundary learning, we propose a Selective Mutual Evolution (SME) module to enable the reciprocal feature learning between them. Then to exploit the global shape context, we propose a Structurally Attentive Refinement (SAR) module to conduct a fine-grained feature refinement for those ambiguous points around the boundary. Finally, to further utilize the multi-scale representation, we integrate the above two modules into a cascaded structure and then introduce a Reciprocal Feature Evolution Network (RFENet) for effective glass-like object segmentation. Extensive experiments demonstrate that our RFENet achieves state-of-the-art performance on three popular public datasets. Code is available at https://github.com/VankouF/RFENet.
Real-world sensors often produce incomplete, irregular, and noisy point clouds, making point cloud completion increasingly important. However, most existing completion methods rely on large paired datasets for training, which is labor-intensive. This paper proposes RaPD, a novel semi-supervised point cloud completion method that reduces the need for paired datasets. RaPD utilizes a two-stage training scheme, where a deep semantic prior is learned in stage 1 from unpaired complete and incomplete point clouds, and a semi-supervised prior distillation process is introduced in stage 2 to train a completion network using only a small number of paired samples. Additionally, a self-supervised completion module is introduced to improve performance using unpaired incomplete point clouds. Experiments on multiple datasets show that RaPD outperforms previous methods in both homologous and heterologous scenarios.
This article proposes a novel spectral domain based solution to the challenging polyp segmentation. The main contribution is based on an interesting finding of the significant existence of the middle frequency sub-band during the CNN process. Consequently, a Sub-Band based Attention (SBA) module is proposed, which uniformly adopts either the high or middle sub-bands of the encoder features to boost the decoder features and thus concretely improve the feature discrimination. A strong encoder supplying informative sub-bands is also very important, while we highly value the local-and-global information enriched CNN features. Therefore, a Transformer Attended Convolution (TAC) module as the main encoder block is introduced. It takes the Transformer features to boost the CNN features with stronger long-range object contexts. The combination of SBA and TAC leads to a novel polyp segmentation framework, SBA-Net. It adopts TAC to effectively obtain encoded features which also input to SBA, so that efficient sub-bands based attention maps can be generated for progressively decoding the bottleneck features. Consequently, SBA-Net can achieve the robust polyp segmentation, as the experimental results demonstrate.
While recent neural image captioning models have shown great promise in terms of automatic metrics, they still struggle with generating generic sentences, which limits their use to only a handful of simple scenarios. On the other hand, negative training has been suggested as an effective way to prevent models from producing frequent yet meaningless sentences. However, when applied to image captioning, this approach may overlook low-frequency but generic and vague sentences, which can be problematic when dealing with diverse and changeable visual scenes. In this paper, we introduce a approach to improve image captioning by integrating negative knowledge that focuses on preventing the model from producing undesirable generic descriptions while addressing previous limitations. We accomplish this by training a negative teacher model that generates image-wise generic sentences with retrieval entropy-filtered data. Subsequently, the student model is required to maximize the distance with multi-level negative knowledge transferring for optimal guiding. Empirical results evaluated on MS COCO benchmark confirm that our plug-and-play framework incorporating unlikely negative knowledge leads to significant improvements in both accuracy and diversity, surpassing previous state-of-the-art methods for distinctive image captioning.
This paper proposes BPNet, a novel end-to-end deep learning framework to learn Bézier 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 Bézier decomposition on NURBS models, we transfer it to guide point cloud segmentation casting off primitive types. A joint optimization framework is proposed to learn Bézier 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.
There are two problems that widely exist in current end-to-end sign language processing architecture. One is the CTC spike phenomenon which weakens the visual representational ability in Continuous Sign Language Recognition (CSLR). The other one is the exposure bias problem which leads to the accumulation of translation errors during inference in Sign Language Translation (SLT). In this paper, we tackle these issues by introducing contrast learning, aiming to enhance both visual-level feature representation and semantic-level error tolerance. Specifically, to alleviate CTC spike phenomenon and enhance visual-level representation, we design a visual contrastive loss by minimizing visual feature distance between different augmented samples of frames in one sign video, so that the model can further explore features by utilizing numerous unlabeled frames in an unsupervised way. To alleviate exposure bias problem and improve semantic-level error tolerance, we design a semantic contrastive loss by re-inputting the predicted sentence into semantic module and comparing features of ground-truth sequence and predicted sequence, for exposing model to its own mistakes. Besides, we propose two new metrics, i.e., Blank Rate and Consecutive Wrong Word Rate to directly reflect our improvement on the two problems. Extensive experimental results on current sign language datasets demonstrate the effectiveness of our approach, which achieves state-of-the-art performance.