论文检索

Han Huang, Yulun Wu, Chao Deng, Ge Gao, Ming Gu, Yu-Shen Liu

Recently, Gaussian Splatting has sparked a new trend in the field of computer vision. Apart from novel view synthesis, it has also been extended to the area of multi-view reconstruction. The latest methods facilitate complete, detailed surface reconstruction while ensuring fast training speed. However, these methods still require dense input views, and their output quality significantly degrades with sparse views. We observed that the Gaussian primitives tend to overfit the few training views, leading to noisy floaters and incomplete reconstruction surfaces. In this paper, we present an innovative sparse-view reconstruction framework that leverages intra-view depth and multi-view feature consistency to achieve remarkably accurate surface reconstruction. Specifically, we utilize monocular depth ranking information to supervise the consistency of depth distribution within patches and employ a smoothness loss to enhance the continuity of the distribution. To achieve finer surface reconstruction, we optimize the absolute position of depth through multi-view projection features. Extensive experiments on DTU and BlendedMVS demonstrate that our method outperforms state-of-the-art methods with a speedup of 60x to 200x, achieving swift and fine-grained mesh reconstruction without the need for costly pre-training.

Dongshuo Huang, Xiaoshui Huang, Chengdong Zhang, Yilei Shi

Recently, there has been considerable exploration of methods for generating 3D point clouds, which is crucial for numerous 3D vision applications. Though conditional generation methods show promising performance, it depends on the additional paired label. On the other hand, unconditional generation methods usually fail to annotate the generated 3D point cloud. In this paper, we introduce a novel self-conditional architecture that trains on unlabeled data and then generates high-quality labeled 3D point clouds. Specifically, we design a module to extract geometry and view features, and then use a feature fusion module to integrate them as a substitute for label embedding in conditional point cloud generation. Then the point cloud generator is trained using the fused features. LPCG also harnesses CLIP to handle the view features of point clouds for generating label information. Besides, we train two feature diffusion modules to capture the essence of multimodal features and obtain diverse fused features for use as conditions in generating point clouds. Experiments on the ShapeNet dataset demonstrate that LPCG achieves state-of-the-art performance for single class generation. Our experimental results show that the accuracy of our generated label annotations reaches around 97.44% for a two-class generation task.

Chihan Huang, Xiaobo Shen

Deep hashing models have achieved great success in retrieval tasks due to their powerful representation and strong information compression capabilities. However, they inherit the vulnerability of deep neural networks to adversarial perturbations. Attackers can severely impact the retrieval capability of hashing models by adding subtle, carefully crafted adversarial perturbations to benign images, transforming them into adversarial images. Most existing adversarial attacks target image classification models, with few focusing on retrieval models. We propose HUANG, the first targeted adversarial attack algorithm to leverage a diffusion model for hashing retrieval in black-box scenarios. In our approach, adversarial denoising uses adversarial perturbations and residual image to guide the shift from benign to adversarial distribution. Extensive experiments demonstrate the superiority of HUANG across different datasets, achieving state-of-the-art performance in black-box targeted attacks. Additionally, the dynamic interplay between denoising and adding adversarial perturbations in adversarial denoising endows HUANG with exceptional robustness and transferability.

Binyuan Huang, Yuqing Wen, Yucheng Zhao, Yaosi Hu, Yingfei Liu, Fan Jia, Weixin Mao, Tiancai Wang, Chi Zhang, Chang Wen Chen 等

Autonomous driving progress relies on large-scale annotated datasets. In this work, we explore the potential of generative models to produce vast quantities of freely-labeled data for autonomous driving applications and present SubjectDrive, the first model proven to scale generative data production in a way that could continuously improve autonomous driving applications. We investigate the impact of scaling up the quantity of generative data on the performance of downstream perception models and find that enhancing data diversity plays a crucial role in effectively scaling generative data production. Therefore, we have developed a novel model equipped with a subject control mechanism, which allows the generative model to leverage diverse external data sources for producing varied and useful data. Extensive evaluations confirm SubjectDrive's efficacy in generating scalable autonomous driving training data, marking a significant step toward revolutionizing data production methods in this field.

Bin Huang, Xin Wang, Hong Chen, Houlun Chen, Yaofei Wu, Wenwu Zhu

Video corpus grounding (VCG), which aims to retrieve relevant video moments from a video corpus, has attracted significant attention in the multimedia research community. However, the existing VCG setting primarily focuses on matching textual descriptions with videos and ignores the distinct visual identities in the videos, thus resulting in inaccurate understanding of video content and deteriorated retrieval performances. To address this limitation, we introduce a novel task, Identity-Text Video Corpus Grounding (ITVCG), which simultaneously utilize textual descriptions and visual identities as queries. As such, ITVCG benefits in enabling more accurate video corpus grounding with visual identities, as well as providing users with more flexible options to locate relevant frames based on either textual descriptions or textual descriptions and visual identities. To conduct evaluations regarding the novel ITVCG task, we propose the TVR-IT dataset, comprising 463 identity images from 6 TV shows, with 68,840 out of 72,840 queries containing at least one identity image. Furthermore, we propose Video-Locator, the first model designed for the ITVCG task. Our proposed Video-Locator integrates video-identity-text alignment and multi-modal fine-grained fusion components, facilitating a video large language model (Video LLM) to jointly understand textual descriptions, visual identities, as well as videos. Experimental results demonstrate the effectiveness of the proposed Video-Locator model and highlight the importance of identity-generalization capability for ITVCG.

Hang Hua, Yunlong Tang, Chenliang Xu, Jiebo Luo

Video summarization aims to create short, accurate, and cohesive summaries of longer videos. Despite the existence of various video summarization datasets, a notable limitation is their limited amount of source videos, which hampers the effective training of advanced large vision-language models (VLMs). Additionally, most existing datasets are created for video-to-video summarization, overlooking the contemporary need for multimodal video content summarization. Recent efforts have been made to expand from unimodal to multimodal video summarization, categorizing the task into three sub-tasks based on the summary's modality: video-to-video (V2V), video-to-text (V2T), and a combination of video and text summarization (V2VT). However, the textual summaries in previous multimodal datasets are inadequate. To address these issues, we introduce Instruct-V2Xum, a cross-modal video summarization dataset featuring 30,000 diverse videos sourced from YouTube, with lengths ranging from 40 to 940 seconds and an average summarization ratio of 16.39%. Each video summary in Instruct-V2Xum is paired with a textual summary that references specific frame indexes, facilitating the generation of aligned video and textual summaries. In addition, we propose a new video summarization framework named V2Xum-LLM. V2Xum-LLM, specifically V2Xum-LLaMA in this study, is the first framework that unifies different video summarization tasks into one large language model's (LLM) text decoder and achieves task-controllable video summarization with temporal prompts and task instructions. Experiments show that V2Xum-LLaMA outperforms strong baseline models on multiple video summarization tasks. Furthermore, we propose an enhanced evaluation metric for V2V and V2VT summarization tasks.

Xiao Hu, Libo Long, Jochen Lang

Dynamic object modeling is a critical challenge in 3D scene reconstruction. Previous methods typically maintain a canonical space to represent the object model, and a deformation field to express the object motion. However, this approach fails when the object undergoes large motions. The position variation caused by significant motion not only complicates the establishment of a canonical space, but also misleads the interpretation of the deformation field. To overcome the above weaknesses, we propose Motion Decoupled Dynamic 3D Gaussian Splatting (M5D-GS), the first 3D-GS model that separates motion and deformation modeling for dynamic object representation with large motion from a monocular camera. M5D-GS increases the practicality of 3D-GS, as it is common for objects to move, rotate, and deform simultaneously. Current datasets only contain object deformations with slight motions. We introduce a pipeline to reuse current benchmarks by adding large motions into the scene. We also introduce a new benchmark featuring several new scenes with complex motions, scenes augmented from previous datasets, and some real world recorded testcases, to fully demonstrate our improvements. Our M5D-GS significantly increases the accuracy under large motion scenarios while maintaining high rendering speed, which makes it suitable for dynamic object representation tasks including 4D novel view synthesis and real-time rendering.

Xiantao Hu, Ying Tai, Xu Zhao, Chen Zhao, Zhenyu Zhang, Jun Li, Bineng Zhong, Jian Yang

Multimodal tracking has garnered widespread attention as a result of its ability to effectively address the inherent limitations of traditional RGB tracking. However, existing multimodal trackers mainly focus on the fusion and enhancement of spatial features or merely leverage the sparse temporal relationships between video frames. These approaches do not fully exploit the temporal correlations in multimodal videos, making it difficult to capture the dynamic changes and motion information of targets in complex scenarios. To alleviate this problem, we propose a unified multimodal spatial-temporal tracking approach named STTrack. In contrast to previous paradigms that solely relied on updating reference information, we introduced a temporal state generator (TSG) that continuously generates a sequence of tokens containing multimodal temporal information. These temporal information tokens are used to guide the localization of the target in the next time state, establish long-range contextual relationships between video frames, and capture the temporal trajectory of the target. Furthermore, at the spatial level, we introduced the mamba fusion and background suppression interactive (BSI) modules. These modules establish a dual-stage mechanism for coordinating information interaction and fusion between modalities. Extensive comparisons on five benchmark datasets illustrate that STTrack achieves state-of-the-art performance across various multimodal tracking scenarios.

Qiang Hu, Zhenyu Yi, Ying Zhou, Fan Huang, Mei Liu, Qiang Li, Zhiwei Wang

We propose MonoBox, an innovative box-supervised segmentation method constrained by monotonicity to liberate its training from the user-unfriendly box-tightness assumption. In contrast to conventional box-supervised segmentation, where the box edges must precisely touch the target boundaries, MonoBox leverages imprecisely-annotated boxes to achieve robust pixel-wise segmentation. The 'linchpin' is that, within the noisy zones around box edges, MonoBox discards the traditional misguiding multiple-instance learning loss, and instead optimizes a carefully-designed objective, termed monotonicity constraint. Along directions transitioning from the foreground to background, this new constraint steers responses to adhere to a trend of monotonically decreasing values. Consequently, the originally unreliable learning within the noisy zones is transformed into a correct and effective monotonicity optimization. Moreover, an adaptive label correction is introduced, enabling MonoBox to enhance the tightness of box annotations using predicted masks from the previous epoch and dynamically shrink the noisy zones as training progresses. We verify MonoBox in the box-supervised segmentation task of polyps, where satisfying box-tightness is challenging due to the vague boundaries between the polyp and normal tissues. Experiments on both public synthetic and in-house real noisy datasets demonstrate that MonoBox exceeds other anti-noise state-of-the-arts by improving Dice by at least 5.5% and 3.3%, respectively.

Qiang Hu, Houqiang Zhong, Zihan Zheng, Xiaoyun Zhang, Zhengxue Cheng, Li Song, Guangtao Zhai, Yanfeng Wang

Neural Radiance Field (NeRF)-based volumetric video has revolutionized visual media by delivering photorealistic Free-Viewpoint Video (FVV) experiences that provide audiences with unprecedented immersion and interactivity. However, the substantial data volumes pose significant challenges for storage and transmission. Existing solutions typically optimize NeRF representation and compression independently or focus on a single fixed rate-distortion (RD) tradeoff. In this paper, we propose VRVVC, a novel end-to-end joint optimization variable-rate framework for volumetric video compression that achieves variable bitrates using a single model while maintaining superior RD performance. Specifically, VRVVC introduces a compact tri-plane implicit residual representation for inter-frame modeling of long-duration dynamic scenes, effectively reducing temporal redundancy. We further propose a variable-rate residual representation compression scheme that leverages a learnable quantization and a tiny MLP-based entropy model. This approach enables variable bitrates through the utilization of predefined Lagrange multipliers to manage the quantization error of all latent representations. Finally, we present an end-to-end progressive training strategy combined with a multi-rate-distortion loss function to optimize the entire framework. Extensive experiments demonstrate that VRVVC achieves a wide range of variable bitrates within a single model and surpasses the RD performance of existing methods across various datasets.

Jintong Hu, Bin Xia, Bin Chen, Wenming Yang, Lei Zhang

Implicit neural representations (INRs) have revolutionized arbitrary-scale super-resolution (ASSR) by modeling images as continuous functions. Most existing INR-based ASSR networks first extract features from the given low-resolution image using an encoder, and then render the super-resolved result via a multi-layer perceptron decoder. Although these approaches have shown promising results, their performance is constrained by the limited representation ability of discrete latent codes in the encoded features. In this paper, we propose a novel ASSR method named GaussianSR that overcomes this limitation through 2D Gaussian Splatting (2DGS). Unlike traditional methods that treat pixels as discrete points, GaussianSR represents each pixel as a continuous Gaussian field. The encoded features are simultaneously refined and upsampled by rendering the mutually stacked Gaussian fields. As a result, long-range dependencies are established to enhance representation ability. In addition, a classifier is developed to dynamically assign Gaussian kernels to all pixels to further improve flexibility. All components of GaussianSR (i.e. encoder, classifier, Gaussian kernels, and decoder) are jointly learned end-to-end. Experiments demonstrate that GaussianSR achieves superior ASSR performance with fewer parameters than existing methods while enjoying interpretable and content-aware feature aggregations.

Teng-Fang Hsiao, Bo-Kai Ruan, Hong-Han Shuai

Painterly image harmonization aims at seamlessly blending disparate visual elements within a single image. However, previous approaches often struggle due to limitations in training data or reliance on additional prompts, leading to inharmonious and content-disrupted output. To surmount these hurdles, we design a Training-and-prompt-Free General Painterly Harmonization method (TF-GPH). TF-GPH incorporates a novel “Similarity Disentangle Mask”, which disentangles the foreground content and background image by redirecting their attention to corresponding reference images, enhancing the attention mechanism for multi-image inputs. Additionally, we propose a “Similarity Reweighting” mechanism to balance harmonization between stylization and content preservation. This mechanism minimizes content disruption by prioritizing the content-similar features within the given background style reference. Finally, we address the deficiencies in existing benchmarks by proposing novel range-based evaluation metrics and a new benchmark to better reflect real-world applications. Extensive experiments demonstrate the efficacy of our method across benchmarks.

Xiaolu Hou, Mingcheng Li, Dingkang Yang, Jiawei Chen, Ziyun Qian, Xiao Zhao, Yue Jiang, Jinjie Wei, Qingyao Xu, Lihua Zhang

With the widespread use of virtual reality applications, 3D scene generation has become a new challenging research frontier. 3D scenes have highly complex structures and need to ensure that the output is dense, coherent, and contains all necessary structures. Many current 3D scene generation methods rely on pre-trained text-to-image diffusion models and monocular depth estimators. However, the generated scenes occupy large amounts of storage space and often lack effective regularisation methods, leading to geometric distortions. To this end, we propose BloomScene, a lightweight structured 3D Gaussian splatting for crossmodal scene generation, which creates diverse and high-quality 3D scenes from text or image inputs. Specifically, a crossmodal progressive scene generation framework is proposed to generate coherent scenes utilizing incremental point cloud reconstruction and 3D Gaussian splatting. Additionally, we propose a hierarchical depth prior-based regularization mechanism that utilizes multi-level constraints on depth accuracy and smoothness to enhance the realism and continuity of the generated scenes. Ultimately, we propose a structured context-guided compression mechanism that exploits structured hash grids to model the context of unorganized anchor attributes, which significantly eliminates structural redundancy and reduces storage overhead. Comprehensive experiments across multiple scenes demonstrate the significant potential and advantages of our framework compared with several baselines.

Wenjin Hou, Dingjie Fu, Kun Li, Shiming Chen, Hehe Fan, Yi Yang

Zero-shot learning (ZSL) aims to recognize unseen classes by transferring semantic knowledge from seen classes to unseen ones, guided by semantic information. To this end, existing works have demonstrated remarkable performance by utilizing global visual features from Convolutional Neural Networks (CNNs) or Vision Transformers (ViTs) for visual-semantic interactions. Due to the limited receptive fields of CNNs and the quadratic complexity of ViTs, however, these visual backbones achieve suboptimal visual-semantic interactions. In this paper, motivated by the visual state space model (i.e., Vision Mamba), which is capable of capturing long-range dependencies and modeling complex visual dynamics, we propose a parameter-efficient ZSL framework called ZeroMamba to advance ZSL. Our ZeroMamba comprises three key components: Semantic-aware Local Projection (SLP), Global Representation Learning (GRL), and Semantic Fusion (SeF). Specifically, SLP integrates semantic embeddings to map visual features to local semantic-related representations, while GRL encourages the model to learn global semantic representations. SeF combines these two semantic representations to enhance the discriminability of semantic features. We incorporate these designs into Vision Mamba, forming an end-to-end ZSL framework. As a result, the learned semantic representations are better suited for classification. Through extensive experiments on four prominent ZSL benchmarks, ZeroMamba demonstrates superior performance, significantly outperforming the state-of-the-art (i.e., CNN-based and ViT-based) methods under both conventional ZSL (CZSL) and generalized ZSL (GZSL) settings.

Shiyu Hou, Tianfei Zhou, Shuai Zhang, Ye Yuan, Guoren Wang

Prompt tuning (PT) has emerged as a key to unlocking the power of visual-language models like CLIP for various downstream tasks. Predominant approaches learn a small set of task-relevant soft prompts by solving an image-class matching problem. Nevertheless, by optimizing merely with respect to class names, they face challenges in learning high performant prompts capable of capturing fine-grained, diverse characteristics of each class, and tends to overfit potentially biased distribution of base classes. In this work, we propose PTinCAS to tackle prompt tuning in a compact attribute space, driven by the premise that attributes offer detailed class interpretations and can facilitate transfer across related categories. Particularly, PTinCAS is grounded in two innovative designs. First, we create a compact attribute space by properly prompting large language models to generate factual descriptions about categories, which are subsequently clustered to form a concise attribute vocabulary. Second, we leverage attributes as a source of supervision in PT to transfer the inherent common sense knowledge in attributes to soft prompts. An object-aware visual prompting mechanism is developed to effortlessly highlight intended regions in the original image, which guides the model towards learning visual attributes associated with object regions rather than the background. We show that PTinCAS not only improves few-shot generalizability compared to existing PT methods, but also provides some level of inherent explainability that helps us understand why a class name is determined based on the attributes activated in an image.

Jie Hou, Jianghong Ma, Xiangyu Mu, Haijun Zhang, Zhao Zhang

The garment structure serves as a crucial medium for expressing the designer's creative vision and showcasing the distinctive character of clothing items. Effective editing of garment structure in fashion images allows for an advanced preview of the design, accelerating the process of garment customization to meet individualized requirements. Although large-scale diffusion models have demonstrated impressive image generation and editing capabilities, no efforts have been made to exploit their potential in part-level editing of images. Unlike previous research, we define a clothing structure editing (CSE) task aimed at accurately editing the local structure of human-centered clothing images through simple instruction-based prompts while maintaining the consistency of clothing appearance. Specifically, this paper develops a new controllable triple-flow framework for structure editing named FashionTailor. An additional network called ClothingNet is proposed to extract the clothing details to address the rigid constraints of the original garment structure. Then, we propose a semantic-refined module to extract the semantic understanding of the source image and adaptively focus on the part to be edited. We also design a cross-blend attention mechanism to integrate fine-grained clothing features to guarantee precise alignment between appearance and target structure features. In addition, a garment structure dataset called StructureFashion has been collated, wherein each item of clothing is represented by multiple photos with diverse structure characteristics, containing over six million pairs. Finally, our method supports editing the structure of multiple parts on a garment simultaneously. Extensive experiments validate the effectiveness of our method for editing part-level human images in StructureFashion dataset and real-scenarios.

Yan Hong, Jianming Feng, Haoxing Chen, Jun Lan, Huijia Zhu, Weiqiang Wang, Jianfu Zhang

The development of text-to-image generative models has enabled the creation of images so realistic that distinguishing between AI-generated images and real photos is becoming a challenge. This progress offers new possibilities but also raises concerns over privacy, authenticity, and security. Detecting AI-generated images is crucial to prevent misuse. To assess the generalizability and robustness of AI-generated image detection, we present a large-scale dataset, referred to as WildFake. This dataset features cutting-edge image generators, a wide variety of generator categories, and generators for various applications, organized in a hierarchical framework. WildFake collects fake images from the open-source community, enriching its diversity with a broad range of image classes and image styles. Its design significantly improves the effectiveness of detection algorithms, making it a valuable resource for enhancing AI-generated image detection in practical applications. Our evaluations offer insights into the performance of generative models at various levels, showcasing WildFake's unique hierarchical structure's benefits.

Cuong Manh Hoang, Yeejin Lee, Byeongkeun Kang

This work addresses the task of generalized class discovery (GCD) in instance segmentation. The goal is to discover novel classes and obtain a model capable of segmenting instances of both known and novel categories, given labeled and unlabeled data. Since the real world contains numerous objects with long-tailed distributions, the instance distribution for each class is inherently imbalanced. To address the imbalanced distributions, we propose an instance-wise temperature assignment (ITA) method for contrastive learning and class-wise reliability criteria for pseudo-labels. The ITA method relaxes instance discrimination for samples belonging to head classes to enhance GCD. The reliability criteria are to avoid excluding most pseudo-labels for tail classes when training an instance segmentation network using pseudo-labels from GCD. Additionally, we propose dynamically adjusting the criteria to leverage diverse samples in the early stages while relying only on reliable pseudo-labels in the later stages. We also introduce an efficient soft attention module to encode object-specific representations for GCD. Finally, we evaluate our proposed method by conducting experiments on two settings: COCO$_{half}$ + LVIS and LVIS + Visual Genome. The experimental results demonstrate that the proposed method outperforms previous state-of-the-art methods.

Miran Heo, Seoung Wug Oh, Seon Joo Kim, Joon-Young Lee

Recent advancements in online Video Instance Segmentation (VIS) methods show notable performance improvements across benchmarks. However, the leading methods in the tracking-by-detection paradigm often result in temporally inconsistent predictions at both instance-level and pixel-level that lead to visually unsatisfactory outcomes. To address these challenges, we propose RoCoVIS, a simple yet effective approach that integrates segmentation and tracking to provide consistent online VIS. Our approach is an end-to-end sequential learning where object queries are propagated through mask predictions, improving the accuracy of temporal instance mapping at the pixel level. Additionally, we propose a new label assignment criterion in harmony with our approach. We also examine the limitations and challenges presented by the current standard evaluation protocol (AP) and suggest adopting additional metrics, Tube-Boundary AP and AP_Pool. RoCoVIS demonstrates superior performance on challenging VIS benchmarks with a Swin-L backbone and shows competitive results when employing a ResNet-50 backbone. By employing Tube-Boundary AP and AP_Pool as metrics to measure mask accuracy and consistency, RoCoVIS outperforms its counterpart, GenVIS, on the HQ-YTVIS and VIPSeg.

Zihao He, Shengchuan Zhang, Runze Hu, Yunhang Shen, Yan Zhang

Super-resolution (SR) techniques are critical for enhancing image quality, particularly in scenarios where high-resolution imagery is essential yet limited by hardware constraints. Existing diffusion models for SR have relied predominantly on Gaussian models for noise generation, which often fall short when dealing with the complex and variable texture inherent in natural scenes. To address these deficiencies, we introduce the Bayesian Uncertainty Guided Diffusion Probabilistic Model (BUFF). BUFF distinguishes itself by incorporating a Bayesian network to generate high-resolution uncertainty masks. These masks guide the diffusion process, allowing for the adjustment of noise intensity in a manner that is both context-aware and adaptive. This novel approach not only enhances the fidelity of super-resolved images to their original high-resolution counterparts but also significantly mitigates artifacts and blurring in areas characterized by complex textures and fine details. The model demonstrates exceptional robustness against complex noise patterns and showcases superior adaptability in handling textures and edges within images. Empirical evidence, supported by visual results, illustrates the model's robustness, especially in challenging scenarios, and its effectiveness in addressing common SR issues such as blurring. Experimental evaluations conducted on the DIV2K dataset reveal that BUFF achieves a notable improvement, with a +0.61 increase compared to baseline in SSIM on BSD100, surpassing traditional diffusion approaches by an average additional +0.20dB PSNR gain. These findings underscore the potential of Bayesian methods in enhancing diffusion processes for SR, paving the way for future advancements in the field.