论文检索

Lingjun Mao, Zineng Tang, Alane Suhr

We study the perception of color illusions by vision-language models. Color illusion, where a person's visual system perceives color differently from actual color, is well-studied in human vision. However, it remains underexplored whether vision-language models (VLMs), trained on large-scale human data, exhibit similar perceptual biases when confronted with such color illusions. We propose an automated framework for generating color illusion images, resulting in RCID (Realistic Color Illusion Dataset), a dataset of 19,000 realistic illusion images. Our experiments show that all studied VLMs exhibit perceptual biases similar human vision. Finally, we train a model to distinguish both human perception and actual pixel differences.

Zhiqi Pang, Junjie Wang, Lingling Zhao, Chunyu Wang

Clothing change person re-identification (CC-ReID) aims to match different images of the same person, even when the clothing varies across images. To reduce manual labeling costs, existing unsupervised CC-ReID methods employ clustering algorithms to generate pseudo-labels. However, they often fail to assign the same pseudo-label to two images with the same identity but different clothing--referred to as a clothing change positive pair--thus hindering clothing-invariant feature learning. To address this issue, we propose the identity-clothing similarity modeling (ICSM) framework. To effectively connect clothing change positive pairs, ICSM first performs clothing-aware learning to leverage all discriminative information, including clothing, to obtain compact clusters. It then extracts cluster-level identity and clothing features and performs inter-cluster similarity estimation to identify clothing change positive clusters, reliable negative clusters, and hard negative clusters for each compact cluster. During optimization, we design an adaptive version of existing optimization methods to enhance similarities of clothing change positive pairs, while also introducing text semantics as a supervisory signal to further promote clothing invariance. Extensive experimental results across multiple datasets validate the effectiveness of the proposed framework, demonstrating its superiority over existing unsupervised methods and its competitiveness with some supervised approaches.

Weitao Feng, Hang Zhou, Jing Liao, Li Cheng, Wenbo Zhou

We present a novel approach for indoor scene synthesis, which learns to arrange decomposed cuboid primitives to represent 3D objects within a scene. Unlike conventional methods that use bounding boxes to determine the placement and scale of 3D objects, our approach leverages cuboids as a straightforward yet highly effective alternative for modeling objects. This allows for compact scene generation while minimizing object intersections. Our approach, coined CASAGPT for Cuboid Arrangement and Scene Assembly, employs an autoregressive model to sequentially arrange cuboids, producing physically plausible scenes. By applying rejection sampling during the fine-tuning stage to filter out scenes with object collisions, our model further reduces intersections and enhances scene quality. Additionally, we introduce a refined dataset, 3DFRONT-NC, which eliminates significant noise presented in the original dataset, 3D-FRONT. Extensive experiments on the 3D-FRONT dataset as well as our dataset demonstrate that our approach consistently outperforms the state-of-the-art methods, enhancing the realism of generated scenes, and providing a promising direction for 3D scene synthesis.

Jiangyi Wang, Na Zhao

Active learning has emerged as a promising approach to reduce the substantial annotation burden in 3D object detection tasks, spurring several initiatives in outdoor environments. However, its application in indoor environments remains unexplored. Compared to outdoor 3D datasets, indoor datasets face significant challenges, including fewer training samples per class, a greater number of classes, more severe class imbalance, and more diverse scene types and intra-class variances.This paper presents the first study on active learning for indoor 3D object detection, where we propose a novel framework tailored for this task. Our method incorporates two key criteria - uncertainty and diversity - to actively select the most ambiguous and informative unlabeled samples for annotation. The uncertainty criterion accounts for both inaccurate detections and undetected objects, ensuring that the most ambiguous samples are prioritized. Meanwhile, the diversity criterion is formulated as a joint optimization problem that maximizes the diversity of both object class distributions and scene types, using a new Class-aware Adaptive Prototype (CAP) bank. The CAP bank dynamically allocates representative prototypes to each class, helping to capture varying intra-class diversity across different categories.We evaluate our method on SUN RGB-D and ScanNetV2, where it outperforms baselines by a significant margin, achieving over 85% of fully-supervised performance with just 10% of the annotation budget.

Jiaqi Liu, Jichao Zhang, Paolo Rota, Nicu Sebe

The Latent Diffusion Model (LDM) has demonstrated strong capabilities in high-resolution image generation and has been widely employed for Pose-Guided Person Image Synthesis (PGPIS), yielding promising results. However, the compression process of LDM often results in the deterioration of details, particularly in sensitive areas such as facial features and clothing textures. In this paper, we propose a Multi-focal Conditioned Latent Diffusion (MCLD) method to address these limitations by conditioning the model on disentangled, pose-invariant features from these sensitive regions. Our approach utilizes a multi-focal condition aggregation module, which effectively integrates facial identity and texture-specific information, enhancing the model's ability to produce appearance realistic and identity-consistent images. Our method demonstrates consistent identity and appearance generation on the DeepFashion dataset and enables flexible person image editing due to its generation consistency. The code is available at https://github.com/jqliu09/mcld.

Xiaohai Li, Bineng Zhong, Qihua Liang, Zhiyi Mo, Jian Nong, Shuxiang Song

The consistency between the semantic information provided by the multi-modal reference and the tracked object is crucial for visual-language (VL) tracking. However, existing VL tracking frameworks rely on static multi-modal references to locate dynamic objects, which can lead to semantic discrepancies and reduce the robustness of the tracker. To address this issue, we propose a novel vision-language tracking framework, named DUTrack, which captures the latest state of the target by dynamically updating multi-modal references to maintain consistency.Specifically, we introduce a Dynamic Language Update Module, which leverages a large language model to generate dynamic language descriptions for the object based on visual features and object category information. Then, we design a Dynamic Template Capture Module, which captures the regions in the image that highly match the dynamic language descriptions. Furthermore, to ensure the efficiency of description generation, we design an update strategy that assesses changes in target displacement, scale, and other factors to decide on updates. Finally, the dynamic template and language descriptions that record the latest state of the target are used to update the multi-modal references, providing more accurate reference information for subsequent inference and enhancing the robustness of the tracker.DUTrack achieves new state-of-the-art performance on four mainstream vision-language and two vision-only tracking benchmarks, including LaSOT, LaSOT_ext, TNL2K, OTB99-Lang, GOT-10K, and UAV123.

Marc-Antoine Lavoie, Anas Mahmoud, Steven L. Waslander

The current state-of-the-art methods in domain adaptive object detection (DAOD) use Mean Teacher self-labelling, where a teacher model, directly derived as an exponential moving average of the student model, is used to generate labels on the target domain which are then used to improve both models in a positive loop. This couples learning and generating labels on the target domain, and other recent works also leverage the generated labels to add additional domain alignment losses. We believe this coupling is brittle and excessively constrained: there is no guarantee that a student trained only on source data can generate accurate target domain labels and initiate the positive feedback loop, and much better target domain labels can likely be generated by using a large pretrained network that has been exposed to much more data. Vision foundational models are exactly such models, and they have shown impressive task generalization capabilities even when frozen. We want to leverage these models for DAOD and introduce DINO Teacher, which consists of two components. First, we train a new labeller on source data only using a large frozen DINOv2 backbone and show it generates more accurate labels than Mean Teacher. Next, we align the student's source and target image patch features with those from a DINO encoder, driving source and target representations closer to the generalizable DINO representation. We obtain state-of-the-art performance on multiple DAOD datasets.

Bo Ji, Angela Yao

Standard 3D Gaussian Splatting (3DGS) relies on known or pre-computed camera poses and a sparse point cloud, obtained from structure-from-motion (SfM) preprocessing, to initialize and grow 3D Gaussians. We propose a novel SfM-Free 3DGS (SFGS) method for video input, eliminating the need for known camera poses and SfM preprocessing. Our approach introduces a hierarchical training strategy that trains and merges multiple 3D Gaussian representations -- each optimized for specific scene regions -- into a single, unified 3DGS model representing the entire scene. To compensate for large camera motions, we leverage video frame interpolation models. Additionally, we incorporate multi-source supervision to reduce overfitting and enhance representation. Experimental results reveal that our approach significantly surpasses state-of-the-art SfM-free novel view synthesis methods. On the Tanks and Temples dataset, we improve PSNR by an average of 2.25dB, with a maximum gain of 3.72dB in the best scene. On the CO3D-V2 dataset, we achieve an average PSNR boost of 1.74dB, with a top gain of 3.90dB. The code is available at \href https://github.com/jibo27/3DGS_Hierarchical_Training/ https://github.com/jibo27/3DGS_Hierarchical_Training .

Ronghao Dang, Yuqian Yuan, Wenqi Zhang, Yifei Xin, Boqiang Zhang, Long Li, Liuyi Wang, Qinyang Zeng, Xin Li, Lidong Bing

The enhancement of generalization in robots by large vision-language models (LVLMs) is increasingly evident. Therefore, the embodied cognitive abilities of LVLMs based on egocentric videos are of great interest. However, current datasets for embodied video question answering lack comprehensive and systematic evaluation frameworks. Critical embodied cognitive issues, such as robotic self-cognition, dynamic scene perception, and hallucination, are rarely addressed. To tackle these challenges, we propose ECBench, a high-quality benchmark designed to systematically evaluate the embodied cognitive abilities of LVLMs. ECBench features a diverse range of scene video sources, open and varied question formats, and 30 dimensions of embodied cognition. To ensure quality, balance, and high visual dependence, ECBench uses class-independent meticulous human annotation and multi-round question screening strategies. Additionally, we introduce ECEval, a comprehensive evaluation system that ensures the fairness and rationality of the indicators. Utilizing ECBench, we conduct extensive evaluations of proprietary, open-source, and task-specific LVLMs. ECBench is pivotal in advancing the embodied cognitive capabilities of LVLMs, laying a solid foundation for developing reliable core models for embodied agents. All data and code is available at https://github.com/Rh-Dang/ECBench.

Guoxing Sun, Rishabh Dabral, Heming Zhu, Pascal Fua, Christian Theobalt, Marc Habermann

Real-time free-view human rendering from sparse-view RGB inputs is a challenging task due to the sensor scarcity and the tight time budget. To ensure efficiency, recent methods leverage 2D CNNs operating in texture space to learn rendering primitives. However, they either jointly learn geometry and appearance, or completely ignore sparse image information for geometry estimation, significantly harming visual quality and robustness to unseen body poses. To address these issues, we present Double Unprojected Textures, which at the core disentangles coarse geometric deformation estimation from appearance synthesis, enabling robust and photorealistic 4K rendering in real-time. Specifically, we first introduce a novel image-conditioned template deformation network, which estimates the coarse deformation of the human template from a first unprojected texture. This updated geometry is then used to apply a second and more accurate texture unprojection. The resulting texture map has fewer artifacts and better alignment with input views, which benefits our learning of finer-level geometry and appearance represented by Gaussian splats. We validate the effectiveness and efficiency of the proposed method in quantitative and qualitative experiments, which significantly surpasses other state-of-the-art methods.

Zilin Wang, Sangwoo Mo, Stella X. Yu, Sima Behpour, Liu Ren

Adaptive categorization of visual scenes is essential for AI agents to handle changing tasks. Unlike fixed common categories for plants or animals, ad-hoc categories, such as things to sell at a garage sale, are created dynamically to achieve specific tasks. We study open ad-hoc categorization, where the goal is to infer novel concepts and categorize images based on a given context, a small set of labeled exemplars, and some unlabeled data. We have two key insights: 1) recognizing ad-hoc categories relies on the same perceptual processes as common categories; 2) novel concepts can be discovered semantically by expanding contextual cues or visually by clustering similar patterns. We propose OAK, a simple model that introduces a single learnable context token into CLIP, trained with CLIP's objective of aligning visual and textual features and GCD's objective of clustering similar images. On Stanford and Clevr-4 datasets, OAK consistently achieves the state-of-art in accuracy and concept discovery across multiple categorizations, including 87.4% novel accuracy on Stanford Mood, surpassing CLIP and GCD by over 50%. Moreover, OAK generates interpretable saliency maps, focusing on hands for Action, faces for Mood, and backgrounds for Location, promoting transparency and trust while enabling accurate and flexible categorization.

Shin-Fang Chng, Hemanth Saratchandran, Simon Lucey

Neural fields encode continuous multidimensional signals as neural networks, enabling diverse applications in computer vision, robotics, and geometry. While Adam is effective for stochastic optimization, it often requires long training times. To address this, we explore alternative optimization techniques to accelerate training without sacrificing accuracy. Traditional second-order methods like L-BFGS are unsuitable for stochastic settings. We propose a theoretical framework for training neural fields with curvature-aware diagonal preconditioners, demonstrating their effectiveness across tasks such as image reconstruction, shape modeling, and Neural Radiance Fields (NeRF).

Tiago Novello, Diana Aldana, Andre Araujo, Luiz Velho

Sinusoidal neural networks have been shown effective as implicit neural representations (INRs) of low-dimensional signals, due to their smoothness and high representation capacity. However, initializing and training them remain empirical tasks which lack on deeper understanding to guide the learning process. To fill this gap, our work introduces a theoretical framework that explains the capacity property of sinusoidal networks and offers robust control mechanisms for initialization and training. Our analysis is based on a novel amplitude-phase expansion of the sinusoidal multilayer perceptron, showing how its layer compositions produce a large number of new frequencies expressed as integer combinations of the input frequencies. This relationship can be directly used to initialize the input neurons, as a form of spectral sampling, and to bound the network's spectrum while training. Our method, referred to as TUNER (TUNing sinusoidal nEtwoRks), greatly improves the stability and convergence of sinusoidal INR training, leading to detailed reconstructions, while preventing overfitting.

Chong Bao, Xiyu Zhang, Zehao Yu, Jiale Shi, Guofeng Zhang, Songyou Peng, Zhaopeng Cui

Neural rendering has demonstrated remarkable success in high-quality 3D neural reconstruction and novel view synthesis with dense input views and accurate poses. However, applying it to sparse, unposed views in unbounded 360* scenes remains a challenging problem. In this paper, we propose a novel neural rendering framework to accomplish the unposed and extremely sparse-view 3D reconstruction in unbounded 360* scenes. To resolve the spatial ambiguity inherent in unbounded scenes with sparse input views, we propose a layered Gaussian-based representation to effectively model the scene with distinct spatial layers. By employing a dense stereo reconstruction model to recover coarse geometry, we introduce a layer-specific bootstrap optimization to refine the noise and fill occluded regions in the reconstruction. Furthermore, we propose an iterative fusion of reconstruction and generation alongside an uncertainty-aware training approach to facilitate mutual conditioning and enhancement between these two processes. Comprehensive experiments show that our approach outperforms existing state-of-the-art methods in terms of rendering quality and surface reconstruction accuracy. Project page: https://zju3dv.github.io/free360/

Quanxing Zha, Xin Liu, Shu-Juan Peng, Yiu-ming Cheung, Xing Xu, Nannan Wang

Can we accurately identify the true correspondences from multimodal datasets containing mismatched data pairs? Existing methods primarily emphasize the similarity matching between the representations of objects across modalities, potentially neglecting the crucial relation consistency within modalities that are particularly important for distinguishing the true and false correspondences. Such an omission often runs the risk of misidentifying negatives as positives, thus leading to unanticipated performance degradation. To address this problem, we propose a general Relation Consistency learning framework, namely ReCon, to accurately discriminate the true correspondences among the multimodal data and thus effectively mitigate the adverse impact caused by mismatches. Specifically, ReCon leverages a novel relation consistency learning to ensure the dual-alignment, respectively of, the cross-modal relation consistency between different modalities and the intra-modal relation consistency within modalities. Thanks to such dual constrains on relations, ReCon significantly enhances its effectiveness for true correspondence discrimination and therefore reliably filters out the mismatched pairs to mitigate the risks of wrong supervisions. Extensive experiments on three widely-used benchmark datasets, including Flickr30K, MS-COCO, and Conceptual Captions, are conducted to demonstrate the effectiveness and superiority of ReCon compared with other SOTAs. The code is available at: https://anonymous.4open.science/r/ReCon-NCL.

Chao Wang, Hehe Fan, Huichen Yang, Sarvnaz Karimi, Lina Yao, Yi Yang

Diffusion-based Text-to-Image (T2I) models have demonstrated significant potential in image restoration. However, existing models continue to grapple with challenges such as complex training and prompt design. We introduce a new perspective for improving image restoration by injecting knowledge from pretrained vision-language models into current T2I models. We empirically show that the degradation and content representations in BLIP-2 can be linearly separated, providing promising degradation guidance for image restoration. Specifically, the Feature Difference Instruction (FDI) is first extracted by Q-Formers through a simple subtraction operation based on reference image pairs. Then, we propose a multi-scale FDI adapter to decouple the degradation style and corrupted artifacts, and inject the styleflow exclusively into specific blocks through adapter-tuning, thereby preventing noise interference and eschewing the need for cumbersome weight retraining. In this way, we can train various task-specific adapters according to different degradations, achieving rich detail enhancement in the restoration results. Furthermore, the proposed FDI adapters have attractive properties of practical value, such as composability and generalization ability for all-in-one and mixed-degradation restoration. Extensive experiments under various settings demonstrate that our method has promising repairing quality over 10 image restoration tasks and a wide range of other applications.

Marzi Heidari, Abdullah Alchihabi, Hao Yan, Yuhong Guo

In this work, we introduce a novel problem setup termed as Heterogeneous Semi-Supervised Learning (HSSL), which presents unique challenges by bridging the semi-supervised learning (SSL) task and the unsupervised domain adaptation (UDA) task, and expanding standard semi-supervised learning to cope with heterogeneous training data. At its core, HSSL aims to learn a prediction model using a combination of labeled and unlabeled training data drawn separately from heterogeneous domains that share a common set of semantic categories; this model is intended to differentiate the semantic categories of test instances sampled from both the labeled and unlabeled domains. In particular, the labeled and unlabeled domains have dissimilar label distributions and class feature distributions. This heterogeneity, coupled with the assorted sources of the test data, introduces significant challenges to standard SSL and UDA methods. Therefore, we propose a novel method, Unified Framework for Heterogeneous Semi-supervised Learning (Uni-HSSL), to address HSSL by directly learning a fine-grained classifier from the heterogeneous data, which adaptively handles the inter-domain heterogeneity while leveraging both the unlabeled data and the inter-domain semantic class relationships for cross-domain knowledge transfer and adaptation. We conduct comprehensive experiments and the experimental results validate the efficacy and superior performance of the proposed Uni-HSSL over state-of-the-art semi-supervised learning and unsupervised domain adaptation methods.

Songpengcheng Xia, Yu Zhang, Zhuo Su, Xiaozheng Zheng, Zheng Lv, Guidong Wang, Yongjie Zhang, Qi Wu, Lei Chu, Ling Pei

Estimating full-body motion using the tracking signals of head and hands from VR devices holds great potential for various applications. However, the sparsity and unique distribution of observations present a significant challenge, resulting in an ill-posed problem with multiple feasible solutions (i.e., hypotheses). This amplifies uncertainty and ambiguity in full-body motion estimation, especially for the lower-body joints. Therefore, we propose a new method, EnvPoser, that employs a two-stage framework to perform full-body motion estimation using sparse tracking signals and pre-scanned environment from VR devices. EnvPoser models the multi-hypothesis nature of human motion through an uncertainty-aware estimation module in the first stage. In the second stage, we refine these multi-hypothesis estimates by integrating semantic and geometric environmental constraints, ensuring that the final motion estimation aligns realistically with both the environmental context and physical interactions.Qualitative and quantitative experiments on two public datasets demonstrate that our method achieves state-of-the-art performance, highlighting significant improvements in human motion estimation within motion-environment interaction scenarios. Project page: https://xspc.github.io/EnvPoser/.

Shanshan Huang, Haoxuan Li, Chunyuan Zheng, Mingyuan Ge, Wei Gao, Lei Wang, Li Liu

Fashion image editing is a valuable tool for designers to convey their creative ideas by visualizing design concepts. With the recent advances in text editing methods, significant progress has been made in fashion image editing. However, they face two key challenges: spurious correlations in training data often induce changes in other areas when editing an area representing the intended editing concept, and these models typically lack the ability to edit multiple concepts simultaneously. To address the above challenges, we propose a novel \underline T ext-driven \underline F ashion \underline I mage edi\underline T ing framework called T-FIT to mitigate the impact of spurious correlation by integrating counterfactual reasoning with compositional concept learning to precisely ensure compositional multi-concept fashion image editing relying solely on text descriptions. Specifically, T-FIT includes three key components. (i) Counterfactual abduction module, which learns an exogenous variable of the source image by a denoising U-Net model. (ii) Concept learning module, which identifies concepts in fashion image editing--such as clothing types and colors and projects a target concept into the space spanned from a series of textual prompts. (iii) Concept composition module, which enables simultaneous adjustments of multiple concepts by aggregating each concept's direction vector obtained from the concept learning module. Extensive experiments show that our method can achieve state-of-the-art performance on various fashion image editing tasks, including single-concept editing (e.g., sleeve length, clothing type) and multi-concept editing (e.g., color & sleeve length).

Ming Li, Jike Zhong, Tianle Chen, Yuxiang Lai, Konstantinos Psounis

Recent studies on large language models (LLMs) and large multimodal models (LMMs) have demonstrated promising skills in various domains including science and mathematics. However, their capability in more challenging and real-world related scenarios like engineering has not been systematically studied. To bridge this gap, we propose EEE-Bench, a multimodal benchmark aimed at assessing LMMs' capabilities in solving practical engineering tasks, using electrical and electronics engineering (EEE) as the testbed. Our benchmark consists of 2860 hand-picked and carefully curated problems spanning 10 essential subdomains such as analog circuits, control systems, etc. Compared to other domains, engineering problems are intrinsically 1) more visually complex and versatile and 2) less deterministic in solutions. Successful solutions to these problems often demand more-than-usual rigorous integration of visual and textual information as models need to understand intricate images like abstract circuits and system diagrams while taking professional instructions. Alongside EEE-Bench, we provide extensive quantitative evaluations, fine-grained analysis, and improvement methods using 17 widely-used open- and closed-sourced LLMs and LMMs and 7 popular prompting techniques. Our results reveal notable deficiencies in current foundation models for EEE, including an average performance ranging from 19.48% to 46.78% and a tendency toward "laziness" in overlooking essential visual context. In summary, we believe EEE-Bench not only reveals some noteworthy limitations of LMMs but also provides a valuable resource for advancing research on their application in practical engineering tasks, driving future improvements in their capability to handle complex, real-world scenarios.