Zechuan Li

Tom Tongjia Chen, Hongshan Yu, Zhengeng Yang et al.

Affordance-Centric Question-driven Task Completion (AQTC) has been proposed to acquire knowledge from videos to furnish users with comprehensive and systematic instructions. However, existing methods have hitherto neglected the necessity of aligning spatiotemporal visual and linguistic signals, as well as the crucial interactional information between humans and objects. To tackle these limitations, we propose to combine large-scale pre-trained vision-language and video-language models, which serve to contribute stable and reliable multimodal data and facilitate effective spatiotemporal visual-textual alignment. Additionally, a novel hand-object-interaction (HOI) aggregation module is proposed which aids in capturing human-object interaction information, thereby further augmenting the capacity to understand the presented scenario. Our method achieved first place in the CVPR'2023 AQTC Challenge, with a Recall@1 score of 78.7\%. The code is available at https://github.com/tomchen-ctj/CVPR23-LOVEU-AQTC.

Shuai Yuan, Hanlin Qin, Renke Kou et al.

We pioneer a learning-based single-point prompt paradigm for infrared small target label generation (IRSTLG) to lobber annotation burdens. Unlike previous clustering-based methods, our intuition is that point-guided mask generation just requires one more prompt than target detection, i.e., IRSTLG can be treated as an infrared small target detection (IRSTD) with the location hint. Therefore, we propose an elegant yet effective Energy-Double-Guided Single-point Prompt (EDGSP) framework, aiming to adeptly transform a coarse IRSTD network into a refined label generation method. Specifically, EDGSP comprises three key modules: 1) target energy initialization (TEI), which establishes a foundational outline to streamline the mapping process for effective shape evolution, 2) double prompt embedding (DPE) for rapidly localizing interesting regions and reinforcing high-resolution individual edges to avoid label adhesion, and 3) bounding box-based matching (BBM) for eliminating false masks via considering comprehensive cluster boundary conditions to obtain a reliable output. In this way, pseudo labels generated by three backbones equipped with our EDGSP achieve 100% object-level probability of detection (Pd) and 0% false-alarm rate (Fa) on SIRST, NUDT-SIRST, and IRSTD-1k datasets, with a pixel-level intersection over union (IoU) improvement of 13.28% over state-of-the-art (SOTA) label generation methods. Further applying our inferred masks to train detection models, EDGSP, for the first time, enables a single-point-generated pseudo mask to surpass the manual labels. Even with coarse single-point annotations, it still achieves 99.5% performance of full labeling. Code is available at https://github.com/xdFai/EDGSP.

Tongjia Chen, Hongshan Yu, Zhengeng Yang et al.

Due to the resource-intensive nature of training vision-language models on expansive video data, a majority of studies have centered on adapting pre-trained image-language models to the video domain. Dominant pipelines propose to tackle the visual discrepancies with additional temporal learners while overlooking the substantial discrepancy for web-scaled descriptive narratives and concise action category names, leading to less distinct semantic space and potential performance limitations. In this work, we prioritize the refinement of text knowledge to facilitate generalizable video recognition. To address the limitations of the less distinct semantic space of category names, we prompt a large language model (LLM) to augment action class names into Spatio-Temporal Descriptors thus bridging the textual discrepancy and serving as a knowledge base for general recognition. Moreover, to assign the best descriptors with different video instances, we propose Optimal Descriptor Solver, forming the video recognition problem as solving the optimal matching flow across frame-level representations and descriptors. Comprehensive evaluations in zero-shot, few-shot, and fully supervised video recognition highlight the effectiveness of our approach. Our best model achieves a state-of-the-art zero-shot accuracy of 75.1% on Kinetics-600.

Shuai Guo, Ao Guo, Junchao Zhao et al.

Object-level 3D reconstruction play important roles across domains such as cultural heritage digitization, industrial manufacturing, and virtual reality. However, existing Gaussian Splatting-based approaches generally rely on full-scene reconstruction, in which substantial redundant background information is introduced, leading to increased computational and storage overhead. To address this limitation, we propose an efficient single-object 3D reconstruction method based on 2D Gaussian Splatting. By directly integrating foreground-background probability cues into Gaussian primitives and dynamically pruning low-probability Gaussians during training, the proposed method fundamentally focuses on an object of interest and improves the memory and computational efficiency. Our pipeline leverages probability masks generated by YOLO and SAM to supervise probabilistic Gaussian attributes, replacing binary masks with continuous probability values to mitigate boundary ambiguity. Additionally, we propose a dual-stage filtering strategy for training's startup to suppress background Gaussians. And, during training, rendered probability masks are conversely employed to refine supervision and enhance boundary consistency across views. Experiments conducted on the MIP-360, T&T, and NVOS datasets demonstrate that our method exhibits strong self-correction capability in the presence of mask errors and achieves reconstruction quality comparable to standard 3DGS approaches, while requiring only approximately 1/10 of their Gaussian amount. These results validate the efficiency and robustness of our method for single-object reconstruction and highlight its potential for applications requiring both high fidelity and computational efficiency.

Zechuan Li, Hongshan Yu, Yihao Ding et al.

We propose GO-N3RDet, a scene-geometry optimized multi-view 3D object detector enhanced by neural radiance fields. The key to accurate 3D object detection is in effective voxel representation. However, due to occlusion and lack of 3D information, constructing 3D features from multi-view 2D images is challenging. Addressing that, we introduce a unique 3D positional information embedded voxel optimization mechanism to fuse multi-view features. To prioritize neural field reconstruction in object regions, we also devise a double importance sampling scheme for the NeRF branch of our detector. We additionally propose an opacity optimization module for precise voxel opacity prediction by enforcing multi-view consistency constraints. Moreover, to further improve voxel density consistency across multiple perspectives, we incorporate ray distance as a weighting factor to minimize cumulative ray errors. Our unique modules synergetically form an end-to-end neural model that establishes new state-of-the-art in NeRF-based multi-view 3D detection, verified with extensive experiments on ScanNet and ARKITScenes. Code will be available at https://github.com/ZechuanLi/GO-N3RDet.

Yan Li, Tianyi Zhang, Zechuan Li et al.

Transformer-based Large Language Models (LLMs) struggle with inputs exceeding their training context window due to positional out-of-distribution (O.O.D.) issues that disrupt attention. Existing solutions, including fine-tuning and training-free methods, face challenges like inefficiency, redundant interpolation, logit outliers, or loss of local positional information. We propose Greedy Attention Logit Interpolation (GALI), a training-free method that improves length extrapolation by greedily reusing pretrained positional intervals and interpolating attention logit to eliminate outliers. GALI achieves stable and superior performance across a wide range of long-context tasks without requiring input-length-specific tuning. Our analysis further reveals that LLMs interpret positional intervals unevenly and that restricting interpolation to narrower ranges improves performance, even on short-context tasks. GALI represents a step toward more robust and generalizable long-text processing in LLMs. Our implementation of GALI, along with the experiments from our paper, is open-sourced at https://github.com/adlnlp/Gali.

Zechuan Li, Hongshan Yu, Yihao Ding et al.

3D Scene Question Answering (3D SQA) represents an interdisciplinary task that integrates 3D visual perception and natural language processing, empowering intelligent agents to comprehend and interact with complex 3D environments. Recent advances in large multimodal modelling have driven the creation of diverse datasets and spurred the development of instruction-tuning and zero-shot methods for 3D SQA. However, this rapid progress introduces challenges, particularly in achieving unified analysis and comparison across datasets and baselines. In this survey, we provide the first comprehensive and systematic review of 3D SQA. We organize existing work from three perspectives: datasets, methodologies, and evaluation metrics. Beyond basic categorization, we identify shared architectural patterns across methods. Our survey further synthesizes core limitations and discusses how current trends, such as instruction tuning, multimodal alignment, and zero-shot, can shape future developments. Finally, we propose a range of promising research directions covering dataset construction, task generalization, interaction modeling, and unified evaluation protocols. This work aims to serve as a foundation for future research and foster progress toward more generalizable and intelligent 3D SQA systems.

Yihao Ding, Soyeon Caren Han, Yanbei Jiang et al.

Domain-specific Visually Rich Document Understanding (VRDU) presents significant challenges due to the complexity and sensitivity of documents in fields such as medicine, finance, and material science. Existing Large (Multimodal) Language Models (LLMs/MLLMs) achieve promising results but face limitations such as hallucinations, inadequate domain adaptation, and reliance on extensive fine-tuning datasets. This paper introduces SynDoc, a novel framework that combines discriminative and generative models to address these challenges. SynDoc employs a robust synthetic data generation workflow, using structural information extraction and domain-specific query generation to produce high-quality annotations. Through adaptive instruction tuning, SynDoc improves the discriminative model's ability to extract domain-specific knowledge. At the same time, a recursive inferencing mechanism iteratively refines the output of both models for stable and accurate predictions. This framework demonstrates scalable, efficient, and precise document understanding and bridges the gap between domain-specific adaptation and general world knowledge for document key information extraction tasks.

Yihao Ding, Siwen Luo, Yue Dai et al.

Visually-Rich Document Understanding (VRDU) has emerged as a critical field, driven by the need to automatically process documents containing complex visual, textual, and layout information. Recently, Multimodal Large Language Models (MLLMs) have shown remarkable potential in this domain, leveraging both Optical Character Recognition (OCR)-dependent and OCR-free frameworks to extract and interpret information in document images. This survey reviews recent advancements in MLLM-based VRDU, highlighting three core components: (1) methods for encoding and fusing textual, visual, and layout features; (2) training paradigms, including pretraining strategies, instruction-response tuning, and the trainability of different model modules; and (3) datasets utilized for pretraining, instruction-tuning, and supervised fine-tuning. Finally, we discuss the challenges and opportunities in this evolving field and propose future directions to advance the efficiency, generalizability, and robustness of VRDU systems.

Yong He, Hongshan Yu, Mingtao Feng et al.

Diffusion probabilistic models are traditionally used to generate colors at fixed pixel positions in 2D images. Building on this, we extend diffusion models to point cloud semantic segmentation, where point positions also remain fixed, and the diffusion model generates point labels instead of colors. To accelerate the denoising process in reverse diffusion, we introduce a noisy label embedding mechanism. This approach integrates semantic information into the noisy label, providing an initial semantic reference that improves the reverse diffusion efficiency. Additionally, we propose a point frequency transformer that enhances the adjustment of high-level context in point clouds. To reduce computational complexity, we introduce the position condition into MLP and propose denoising PointNet to process the high-resolution point cloud without sacrificing geometric details. Finally, we integrate the proposed noisy label embedding, point frequency transformer and denoising PointNet in our proposed dual conditional diffusion model-based network (PointDiffuse) to perform large-scale point cloud semantic segmentation. Extensive experiments on five benchmarks demonstrate the superiority of PointDiffuse, achieving the state-of-the-art mIoU of 74.2\% on S3DIS Area 5, 81.2\% on S3DIS 6-fold and 64.8\% on SWAN dataset.