The diversity of tables makes table detection a great challenge, leading to existing models becoming more tedious and complex. Despite achieving high performance, they often overfit to the table style in training set, and suffer from significant performance degradation when encountering out-of-distribution tables in other domains. To tackle this problem, we start from the essence of the table, which is a set of text arranged in rows and columns. Based on this, we propose a novel, light-weighted and robust Table Detection method based on Learning Text Arrangement, namely TDeLTA. TDeLTA takes the text blocks as input, and then models the arrangement of them with a sequential encoder and an attention module. To locate the tables precisely, we design a text-classification task, classifying the text blocks into 4 categories according to their semantic roles in the tables. Experiments are conducted on both the text blocks parsed from PDF and extracted by open-source OCR tools, respectively. Compared to several state-of-the-art methods, TDeLTA achieves competitive results with only 3.1M model parameters on the large-scale public datasets. Moreover, when faced with the cross-domain data under the 0-shot setting, TDeLTA outperforms baselines by a large margin of nearly 7%, which shows the strong robustness and transferability of the proposed model.
论文检索
输入标题、作者或关键词，从 2,866 篇学术成果中精准定位
The choice of the representations is essential for deep gait recognition methods. The binary silhouettes and skeletal coordinates are two dominant representations in recent literature, achieving remarkable advances in many scenarios. However, inherent challenges remain, in which silhouettes are not always guaranteed in unconstrained scenes, and structural cues have not been fully utilized from skeletons. In this paper, we introduce a novel skeletal gait representation named skeleton map, together with SkeletonGait, a skeleton-based method to exploit structural information from human skeleton maps. Specifically, the skeleton map represents the coordinates of human joints as a heatmap with Gaussian approximation, exhibiting a silhouette-like image devoid of exact body structure. Beyond achieving state-of-the-art performances over five popular gait datasets, more importantly, SkeletonGait uncovers novel insights about how important structural features are in describing gait and when they play a role. Furthermore, we propose a multi-branch architecture, named SkeletonGait++, to make use of complementary features from both skeletons and silhouettes. Experiments indicate that SkeletonGait++ outperforms existing state-of-the-art methods by a significant margin in various scenarios. For instance, it achieves an impressive rank-1 accuracy of over 85% on the challenging GREW dataset. The source code is available at https://github.com/ShiqiYu/OpenGait.
In few-shot image generation, directly training GAN models on just a handful of images faces the risk of overfitting. A popular solution is to transfer the models pretrained on large source domains to small target ones. In this work, we introduce WeditGAN, which realizes model transfer by editing the intermediate latent codes w in StyleGANs with learned constant offsets (delta w), discovering and constructing target latent spaces via simply relocating the distribution of source latent spaces. The established one-to-one mapping between latent spaces can naturally prevents mode collapse and overfitting. Besides, we also propose variants of WeditGAN to further enhance the relocation process by regularizing the direction or finetuning the intensity of delta w. Experiments on a collection of widely used source/target datasets manifest the capability of WeditGAN in generating realistic and diverse images, which is simple yet highly effective in the research area of few-shot image generation. Codes are available at https://github.com/Ldhlwh/WeditGAN.
Consistent editing of real images is a challenging task, as it requires performing non-rigid edits (e.g., changing postures) to the main objects in the input image without changing their identity or attributes. To guarantee consistent attributes, some existing methods fine-tune the entire model or the textual embedding for structural consistency, but they are time-consuming and fail to perform non-rigid edits. Other works are tuning-free, but their performances are weakened by the quality of Denoising Diffusion Implicit Model (DDIM) reconstruction, which often fails in real-world scenarios. In this paper, we present a novel approach called Tuning-free Inversion-enhanced Control (TIC), which directly correlates features from the inversion process with those from the sampling process to mitigate the inconsistency in DDIM reconstruction. Specifically, our method effectively obtains inversion features from the key and value features in the self-attention layers, and enhances the sampling process by these inversion features, thus achieving accurate reconstruction and content-consistent editing. To extend the applicability of our method to general editing scenarios, we also propose a mask-guided attention concatenation strategy that combines contents from both the inversion and the naive DDIM editing processes. Experiments show that the proposed method outperforms previous works in reconstruction and consistent editing, and produces impressive results in various settings.
Point cloud completion aims at completing shapes from their partial. Most existing methods utilized shape’s priors information for point cloud completion, such as inputting the partial and getting the complete one through an encoder-decoder deep learning structure. However, it is very often to easily cause the loss of information in the generation process because of the invisibility of missing areas. Unlike most existing methods directly inferring the missing points using shape priors, we address it as a cross-modality task. We propose a new Cross-modal Dual Phases Network (CDPNet) for shape completion. Our key idea is that the global information of the shape is obtained from the extra single-view image, and the partial point clouds provide the geometric information. After that, the multi-modal features jointly guide the specific structural information. To learn the geometric details of the shape, we chose to use patches to preserve the local geometric feature. In this way, we can generate shapes with enough geometric details. Experimental results show that our method achieves state-of-the-art performance on point cloud completion.
Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning
PDF ↗Recently, arbitrary-scale point cloud upsampling mechanism became increasingly popular due to its efficiency and convenience for practical applications. To achieve this, most previous approaches formulate it as a problem of surface approximation and employ point-based networks to learn surface representations. However, learning surfaces from sparse point clouds is more challenging, and thus they often suffer from the low-fidelity geometry approximation. To address it, we propose an arbitrary-scale Point cloud Upsampling framework using Voxel-based Network (PU-VoxelNet). Thanks to the completeness and regularity inherited from the voxel representation, voxel-based networks are capable of providing predefined grid space to approximate 3D surface, and an arbitrary number of points can be reconstructed according to the predicted density distribution within each grid cell. However, we investigate the inaccurate grid sampling caused by imprecise density predictions. To address this issue, a density-guided grid resampling method is developed to generate high-fidelity points while effectively avoiding sampling outliers. Further, to improve the fine-grained details, we present an auxiliary training supervision to enforce the latent geometric consistency among local surface patches. Extensive experiments indicate the proposed approach outperforms the state-of-the-art approaches not only in terms of fixed upsampling rates but also for arbitrary-scale upsampling. The code is available at https://github.com/hikvision-research/3DVision
Image-based virtual try-on aims to transfer a target clothing onto a specific person. A significant challenge is arbitrarily matched clothing and person lack corresponding ground truth to supervised learning. A recent pioneering work leveraged an improved cycleGAN to enable one network to generate the desired image for another network during training. However, there is no difference in the result distribution before and after the clothing changes. Therefore, using two different networks is unnecessary and may even increase the difficulty of convergence. Furthermore, the introduced human parsing used to provide body structure information in the input also have a negative impact on the try-on result. How to employ a single network for supervised learning while eliminating human parsing? To tackle these issues, we present a Cycle mapping Virtual Try-On Network (CycleVTON), which can produce photo-realistic try-on results by using a cycle mapping framework without the parser. In particular, we introduce a flow constraint loss to achieve supervised learning of arbitrarily matched clothing and person as inputs to the deformer, thus naturally mimicking the interaction between clothing and the human body. Additionally, we design a skin generation strategy that can adapt to the shape of the target clothing by dynamically adjusting the skin region, i.e., by first removing and then filling skin areas. Extensive experiments conducted on challenging benchmarks demonstrate that our proposed method exhibits superior performance compared to state-of-the-art methods.
PPEA-Depth: Progressive Parameter-Efficient Adaptation for Self-Supervised Monocular Depth Estimation
PDF ↗Self-supervised monocular depth estimation is of significant importance with applications spanning across autonomous driving and robotics. However, the reliance on self-supervision introduces a strong static-scene assumption, thereby posing challenges in achieving optimal performance in dynamic scenes, which are prevalent in most real-world situations. To address these issues, we propose PPEA-Depth, a Progressive Parameter-Efficient Adaptation approach to transfer a pre-trained image model for self-supervised depth estimation. The training comprises two sequential stages: an initial phase trained on a dataset primarily composed of static scenes, succeeded by an expansion to more intricate datasets involving dynamic scenes. To facilitate this process, we design compact encoder and decoder adapters to enable parameter-efficient tuning, allowing the network to adapt effectively. They not only uphold generalized patterns from pre-trained image models but also retain knowledge gained from the preceding phase into the subsequent one. Extensive experiments demonstrate that PPEA-Depth achieves state-of-the-art performance on KITTI, CityScapes and DDAD datasets.
HybridGait: A Benchmark for Spatial-Temporal Cloth-Changing Gait Recognition with Hybrid Explorations
PDF ↗Existing gait recognition benchmarks mostly include minor clothing variations in the laboratory environments, but lack persistent changes in appearance over time and space. In this paper, we propose the first in-the-wild benchmark CCGait for cloth-changing gait recognition, which incorporates diverse clothing changes, indoor and outdoor scenes, and multi-modal statistics over 92 days. To further address the coupling effect of clothing and viewpoint variations, we propose a hybrid approach HybridGait that exploits both temporal dynamics and the projected 2D information of 3D human meshes. Specifically, we introduce a Canonical Alignment Spatial-Temporal Transformer (CA-STT) module to encode human joint position-aware features, and fully exploit 3D dense priors via a Silhouette-guided Deformation with 3D-2D Appearance Projection (SilD) strategy. Our contributions are twofold: we provide a challenging benchmark CCGait that captures realistic appearance changes over expanded time and space, and we propose a hybrid framework HybridGait that outperforms prior works on CCGait and Gait3D benchmarks. Our project page is available at https://github.com/HCVLab/HybridGait.
Digital image enhancement aims to deliver visually striking, pleasing images that align with human perception. While global techniques can elevate the image's overall aesthetics, fine-grained color enhancement can further boost visual appeal and expressiveness. However, colorists frequently face challenges in achieving accurate, localized color adjustments. Direct composition of these local edits can result in spatial color inconsistencies. Existing methods, including color style transfer and image harmonization, exhibit inconsistencies, especially at boundary regions. Addressing this, we present ChromaFusionNet (CFNet), a novel approach that views the color fusion problem through the lens of image color inpainting. Built on the Vision Transformer architecture, CFNet captures global context and delivers high-fidelity outputs, seamlessly blending colors while preserving boundary integrity. Empirical studies on ImageNet and COCO datasets demonstrate CFNet's superiority over existing methods in maintaining color harmony and color fidelity. Robustness evaluations and user studies have further validated the effectiveness of CFNet. In conclusion, CFNet introduces an innovative approach to seamless, fine-grained color fusion, paving the way for advancements in the domain of fine-grained color editing. Code and pretrained models are available at our project page: https://yidong.pro/projects/cfnet.
As a neuromorphic camera with high temporal resolution, spike camera can capture dynamic scenes with high-speed motion. Recently, spike camera with a color filter array (CFA) has been developed for color imaging. There are some methods for spike camera demosaicing to reconstruct color images from Bayer-pattern spike streams. However, the demosaicing results are bothered by severe noise in spike streams, to which previous works pay less attention. In this paper, we propose an iterative joint demosaicing and denoising network (SJDD-Net) for spike cameras based on the observation model. Firstly, we design a color spike representation (CSR) to learn latent representation from Bayer-pattern spike streams. In CSR, we propose an offset-sharing deformable convolution module to align temporal features of color channels. Then we develop a spike noise estimator (SNE) to obtain features of the noise distribution. Finally, a color correlation prior (CCP) module is proposed to utilize the color correlation for better details. For training and evaluation, we designed a spike camera simulator to generate Bayer-pattern spike streams with synthesized noise. Besides, we captured some Bayer-pattern spike streams, building the first real-world captured dataset to our knowledge. Experimental results show that our method can restore clean images from Bayer-pattern spike streams. The source codes and dataset are available at https://github.com/csycdong/SJDD-Net.
Learning Multi-Modal Cross-Scale Deformable Transformer Network for Unregistered Hyperspectral Image Super-resolution
PDF ↗Hyperspectral image super-resolution (HSI-SR) is a technology to improve the spatial resolution of HSI. Existing fusion-based SR methods have shown great performance, but still have some problems as follows: 1) existing methods assume that the auxiliary image providing spatial information is strictly registered with the HSI, but images are difficult to be registered finely due to the shooting platforms, shooting viewpoints and the influence of atmospheric turbulence; 2) most of the methods are based on convolutional neural networks (CNNs), which is effective for local features but cannot utilize the global features. To this end, we propose a multi-modal cross-scale deformable transformer network (M2DTN) to achieve unregistered HSI-SR. Specifically, we formulate a spectrum-preserving based spatial-guided registration-SR unified model (SSRU) from the view of the realistic degradation scenarios. According to SSRU, we propose multi-modal registration deformable module (MMRD) to align features between different modalities by deformation field. In order to efficiently utilize the unique information between different modals, we design multi-scale feature transformer (MSFT) to emphasize the spatial-spectral features at different scales. In addition, we propose the cross-scale feature aggregation module (CSFA) to accurately reconstruct the HSI by aggregating feature information at different scales. Experiments show that M2DTN outperforms the-state-of-the-art HSI-SR methods. Code is obtainable at https://github.com/Jiahuiqu/M2DTN.
Car detection is an important task that serves as a crucial prerequisite for many automated driving functions. The large variations in lighting/weather conditions and vehicle densities of the scenes pose significant challenges to existing car detection algorithms to meet the highly accurate perception demand for safety, due to the unstable/limited color information, which impedes the extraction of meaningful/discriminative features of cars. In this work, we present a novel learning-based car detection method that leverages trichromatic linear polarization as an additional cue to disambiguate such challenging cases. A key observation is that polarization, characteristic of the light wave, can robustly describe intrinsic physical properties of the scene objects in various imaging conditions and is strongly linked to the nature of materials for cars (e.g., metal and glass) and their surrounding environment (e.g., soil and trees), thereby providing reliable and discriminative features for robust car detection in challenging scenes. To exploit polarization cues, we first construct a pixel-aligned RGB-Polarization car detection dataset, which we subsequently employ to train a novel multimodal fusion network. Our car detection network dynamically integrates RGB and polarization features in a request-and-complement manner and can explore the intrinsic material properties of cars across all learning samples. We extensively validate our method and demonstrate that it outperforms state-of-the-art detection methods. Experimental results show that polarization is a powerful cue for car detection. Our code is available at https://github.com/wind1117/AAAI24-PCDNet.
Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection
PDF ↗Open World Object Detection (OWOD) is a challenging and realistic task that extends beyond the scope of standard Object Detection task. It involves detecting both known and unknown objects while integrating learned knowledge for future tasks. However, the level of "unknownness" varies significantly depending on the context. For example, a tree is typically considered part of the background in a self-driving scene, but it may be significant in a household context. We argue that this contextual information should already be embedded within the known classes. In other words, there should be a semantic or latent structure relationship between the known and unknown items to be discovered. Motivated by this observation, we propose Hyp-OW, a method that learns and models hierarchical representation of known items through a SuperClass Regularizer. Leveraging this representation allows us to effectively detect unknown objects using a similarity distance-based relabeling module. Extensive experiments on benchmark datasets demonstrate the effectiveness of Hyp-OW, achieving improvement in both known and unknown detection (up to 6 percent). These findings are particularly pronounced in our newly designed benchmark, where a strong hierarchical structure exists between known and unknown objects.
Transferable Adversarial Attacks for Object Detection Using Object-Aware Significant Feature Distortion
PDF ↗Transferable black-box adversarial attacks against classifiers by disturbing the intermediate-layer features have been extensively studied in recent years. However, these methods have not yet achieved satisfactory performances when directly applied to object detectors. This is largely because the features of detectors are fundamentally different from that of the classifiers. In this study, we propose a simple but effective method to improve the transferability of adversarial examples for object detectors by leveraging the properties of spatial consistency and limited equivariance of object detectors’ features. Specifically, we combine a novel loss function and deliberately designed data augmentation to distort the backbone features of object detectors by suppressing significant features corresponding to objects and amplifying the surrounding vicinal features corresponding to object boundaries. As such the target object and background area on the generated adversarial samples are more likely to be confused by other detectors. Extensive experimental results show that our proposed method achieves state-of-the-art black-box transferability for untargeted attacks on various models, including one/two-stage, CNN/Transformer-based, and anchor-free/anchor-based detectors.
Human motion forecasting, with the goal of estimating future human behavior over a period of time, is a fundamental task in many real-world applications. However, existing works typically concentrate on foretelling the major joints of the human body without considering the delicate movements of the human hands. In practical applications, hand gesture plays an important role in human communication with the real world, and expresses the primary intention of human beings. In this work, we are the first to formulate whole-body human pose forecasting task, which jointly predicts future both body and gesture activities. Correspondingly, we propose a novel Encoding-Alignment-Interaction (EAI) framework that aims to predict both coarse (body joints) and fine-grained (gestures) activities collaboratively, enabling expressive and cross-facilitated forecasting of 3D whole-body human motions. Specifically, our model involves two key constituents: cross-context alignment (XCA) and cross-context interaction (XCI). Considering the heterogeneous information within the whole-body, XCA aims to align the latent features of various human components, while XCI focuses on effectively capturing the context interaction among the human components. We conduct extensive experiments on a newly-introduced large-scale benchmark and achieve state-of-the-art performance. The code is public for research purposes at https://github.com/Dingpx/EAI.
Weak Distribution Detectors Lead to Stronger Generalizability of Vision-Language Prompt Tuning
PDF ↗We propose a generalized method for boosting the generalization ability of pre-trained vision-language models (VLMs) while fine-tuning on downstream few-shot tasks. The idea is realized by exploiting out-of-distribution (OOD) detection to predict whether a sample belongs to a base distribution or a novel distribution and then using the score generated by a dedicated competition based scoring function to fuse the zero-shot and few-shot classifier. The fused classifier is dynamic, which will bias towards the zero-shot classifier if a sample is more likely from the distribution pre-trained on, leading to improved base-to-novel generalization ability. Our method is performed only in test stage, which is applicable to boost existing methods without time-consuming re-training. Extensive experiments show that even weak distribution detectors can still improve VLMs' generalization ability. Specifically, with the help of OOD detectors, the harmonic mean of CoOp and ProGrad increase by 2.6 and 1.5 percentage points over 11 recognition datasets in the base-to-novel setting.
With the rapid development of 3D movie and light-field displays, there is a growing demand for stereo videos. However, generating high-quality stereo videos from planar videos remains a challenging task. Traditional depth-image-based rendering techniques struggle to effectively handle the problem of occlusion exposure, which occurs when the occluded contents become visible in other views. Recently, the single-view multiplane images (MPI) representation has shown promising performance for planar video stereoscopy. However, the MPI still lacks real details that are occluded in the current frame, resulting in blurry artifacts in occlusion exposure regions. In fact, planar videos can leverage complementary information from adjacent frames to predict a more complete scene representation for the current frame. Therefore, this paper extends the MPI from still frames to the temporal domain, introducing the temporal MPI (TMPI). By extracting complementary information from adjacent frames based on optical flow guidance, obscured regions in the current frame can be effectively repaired. Additionally, a new module called masked optical flow warping (MOFW) is introduced to improve the propagation of pixels along optical flow trajectories. Experimental results demonstrate that the proposed method can generate high-quality stereoscopic or light-field videos from a single view and reproduce better occluded details than other state-of-the-art (SOTA) methods. https://github.com/Dio3ding/TMPI
Rescaling the backpropagated gradient of contrastive loss has made significant progress in descriptor learning. However, current gradient modulation strategies have no regard for the varying distribution of global gradients, so they would suffer from changes in training phases or datasets. In this paper, we propose a dynamic gradient modulation, named SDGMNet, for contrastive local descriptor learning. The core of our method is formulating modulation functions with dynamically estimated statistical characteristics. Firstly, we introduce angle for distance measure after deep analysis on backpropagation of pair-wise loss. On this basis, auto-focus modulation is employed to moderate the impact of statistically uncommon individual pairs in stochastic gradient descent optimization; probabilistic margin cuts off the gradients of proportional triplets that have achieved enough optimization; power adjustment balances the total weights of negative pairs and positive pairs. Extensive experiments demonstrate that our novel descriptor surpasses previous state-of-the-art methods in several tasks including patch verification, retrieval, pose estimation, and 3D reconstruction.
Attention-based graph neural networks have made great progress in feature matching. However, the literature lacks a comprehensive understanding of how the attention mechanism operates for feature matching. In this paper, we rethink cross- and self-attention from the viewpoint of traditional feature matching and filtering. To facilitate the learning of matching and filtering, we incorporate the similarity of descriptors into cross-attention and relative positions into self-attention. In this way, the attention can concentrate on learning residual matching and filtering functions with reference to the basic functions of measuring visual and spatial correlation. Moreover, we leverage descriptor similarity and relative positions to extract inter- and intra-neighbors. Then sparse attention for each point can be performed only within its neighborhoods to acquire higher computation efficiency. Extensive experiments, including feature matching, pose estimation and visual localization, confirm the superiority of the proposed method. Our codes are available at https://github.com/ACuOoOoO/ResMatch.