AAAI 2025 论文列表

Video summarization aims to eliminate visual redundancy while retaining key parts of video to construct concise and comprehensive synopses. Most existing methods use discriminative models to predict the importance scores of video frames. However, these methods are susceptible to annotation inconsistency caused by the inherent subjectivity of different annotators when annotating the same video. In this paper, we introduce a generative framework for video summarization that learns how to generate summaries from a probability distribution perspective, effectively reducing the interference of subjective annotation noise. Specifically, we propose a novel diffusion summarization method based on the Denoising Diffusion Probabilistic Model (DDPM), which learns the probability distribution of training data through noise prediction, and generates summaries by iterative denoising. Our method is more resistant to subjective annotation noise, and is less prone to overfitting the training data than discriminative methods, with strong generalization ability. Moreover, to facilitate training DDPM with limited data, we employ an unsupervised video summarization model to implement the earlier denoising process. Extensive experiments on various datasets (TVSum, SumMe, and FPVSum) demonstrate the effectiveness of our method.

The demand for producing short-form videos for sharing on social media platforms has experienced significant growth in recent times. Despite notable advancements in the fields of video summarization and highlight detection, which can create partially usable short films from raw videos, these approaches are often domain-specific and require an in-depth understanding of real-world video content. To tackle this predicament, we propose Repurpose-10K, an extensive dataset comprising over 10,000 videos with more than 120,000 annotated clips aimed at resolving the video long-to-short task. Recognizing the inherent constraints posed by untrained human annotators, which can result in inaccurate annotations for repurposed videos, we propose a two-stage solution to obtain annotations from real-world user-generated content. Furthermore, we offer a baseline model to address this challenging task by integrating audio, visual, and caption aspects through a cross-modal fusion and alignment framework. We aspire for our work to ignite groundbreaking research in the lesser-explored realms of video repurposing.

Propagation-based video inpainting using optical flow at the pixel or feature level has recently garnered significant attention. However, it has limitations such as the inaccuracy of optical flow prediction and the propagation of noise over time. These issues result in non-uniform noise and time consistency problems throughout the video, which are particularly pronounced when the removed area is large and involves substantial movement. To address these issues, we propose a novel First Frame Filling Video Diffusion Inpainting model (FFF-VDI). We design FFF-VDI inspired by the capabilities of pre-trained image-to-video diffusion models that can transform the first frame image into a highly natural video. To apply this to the video inpainting task, we propagate the noise latent information of future frames to fill the masked areas of the first frame's noise latent code. Next, we fine-tune the pre-trained image-to-video diffusion model to generate the inpainted video. The proposed model addresses the limitations of existing methods that rely on optical flow quality, producing much more natural and temporally consistent videos. This proposed approach is the first to effectively integrate image-to-video diffusion models into video inpainting tasks. Through various comparative experiments, we demonstrate that the proposed model can robustly handle diverse inpainting types with high quality.

A recent endeavor in one class of video anomaly detection is to leverage diffusion models and posit the task as a generation problem, where the diffusion model is trained to recover normal patterns exclusively, thus reporting abnormal patterns as outliers. Yet, existing attempts neglect the various formations of anomaly and predict normal samples at the feature level regardless that abnormal objects in surveillance videos are often relatively small. To address this, a novel patch-based diffusion model is proposed, specifically engineered to capture fine-grained local information. We further observe that anomalies in videos manifest themselves as deviations in both appearance and motion. Therefore, we argue that a comprehensive solution must consider both of these aspects simultaneously to achieve accurate frame prediction. To address this, we introduce innovative motion and appearance conditions that are seamlessly integrated into our patch diffusion model. These conditions are designed to guide the model in generating coherent and contextually appropriate predictions for both semantic content and motion relations. Experimental results on four challenging video anomaly detection datasets empirically substantiate the efficacy of our proposed approach, demonstrating that it consistently outperforms most existing methods in detecting abnormal behaviors.

Understanding internal joint loading is critical for diagnosing gait-related diseases such as knee osteoarthritis; however, current methods of measuring joint risk factors are time-consuming, expensive, and restricted to lab settings. In this paper, we enable the large-scale, cost-effective biomechanical analysis of joint loading via three key contributions: the development and deployment of novel instrumented insoles, the creation of a large multimodal biomechanics dataset (VidSole), and a baseline deep learning pipeline to predict internal joint loading factors. Our novel instrumented insole measures the tri-axial forces and moments across five high-pressure points under the foot. VidSole consists of the forces and moments measured by these insoles along with corresponding RGB video from two viewpoints, 3D body motion capture, and force plate data for over 2,600 trials of 52 diverse participants performing four fundamental activities of daily living (sit-to-stand, stand-to-sit, walking, and running). We feed the insole data and kinematic parameters extractable from video (i.e., pose, knee angle) into a deep learning pipeline consisting of an ensemble Gated Recurrent Unit (GRU) activity classifier followed by activity-specific Long Short Term Memory (LSTM) regression networks to estimate knee adduction moment (KAM), a biomechanical risk factor for knee osteoarthritis. The successful classification of activities at an accuracy of 99.02 percent and KAM estimation with mean absolute error (MAE) less than 0.5 percent*body weight*height, the current threshold for accurately detecting knee osteoarthritis with KAM, illustrates the usefulness of our dataset for future research and clinical settings.

Despite the significant impact of visual events on human cognition, understanding events in videos remains a challenging task for AI due to their complex structures, semantic hierarchies, and dynamic evolution. To address this, we propose the task of video event understanding that extracts event scripts and makes predictions with these scripts from videos. To support this task, we introduce VidEvent, a large-scale dataset containing over 23,000 well-labeled events, featuring detailed event structures, broad hierarchies, and logical relations extracted from movie recap videos. The dataset was created through a meticulous annotation process, ensuring high-quality and reliable event data. We also provide comprehensive baseline models offering detailed descriptions of their architecture and performance metrics. These models serve as benchmarks for future research, facilitating comparisons and improvements. Our analysis of VidEvent and the baseline models highlights the dataset's potential to advance video event understanding and encourages the exploration of innovative algorithms and models.

Despite the advancements of Video Large Language Models (VideoLLMs) in various tasks, they struggle with fine-grained temporal understanding, such as Dense Video Captioning (DVC). DVC is a complicated task of describing all events within a video while also temporally localizing them, which integrates multiple fine-grained tasks, including video segmentation, video captioning, and temporal video grounding. Previous VideoLLMs attempt to solve DVC in a single step, failing to utilize their reasoning capability. Moreover, previous training objectives for VideoLLMs do not fully reflect the evaluation metrics, therefore not providing supervision directly aligned to target tasks. To address such a problem, we propose a novel framework named VidChain comprised of Chain-of-Tasks (CoTasks) and Metric-based Direct Preference Optimization (M-DPO). CoTasks decompose a complex task into a sequence of sub-tasks, allowing VideoLLMs to leverage their reasoning capabilities more effectively. M-DPO aligns a VideoLLM with evaluation metrics, providing fine-grained supervision to each task that is well-aligned with metrics. Applied to two different VideoLLMs, VidChain consistently improves their fine-grained video understanding, thereby outperforming previous VideoLLMs on two different DVC benchmarks and also on the temporal video grounding task.

Inferring the 3D structure of a scene from a single image is an ill-posed and challenging problem in the field of vision-centric autonomous driving. Existing methods usually employ neural radiance fields to produce voxelized 3D occupancy, lacking instance-level semantic reasoning and temporal photometric consistency. In this paper, we propose ViPOcc, which leverages the visual priors from vision foundation models (VFMs) for fine-grained 3D occupancy prediction. Unlike previous works that solely employ volume rendering for RGB and depth image reconstruction, we introduce a metric depth estimation branch, in which an inverse depth alignment module is proposed to bridge the domain gap in depth distribution between VFM predictions and the ground truth. The recovered metric depth is then utilized in temporal photometric alignment and spatial geometric alignment to ensure accurate and consistent 3D occupancy prediction. Additionally, we also propose a semantic-guided non-overlapping Gaussian mixture sampler for efficient, instance-aware ray sampling, which addresses the redundant and imbalanced sampling issue that still exists in previous state-of-the-art methods. Extensive experiments demonstrate the superior performance of ViPOcc in both 3D occupancy prediction and depth estimation tasks on diverse public datasets.

Recent lightweight image captioning models using retrieved data mainly focus on text prompts. However, previous works only utilize the retrieved text as text prompts, and the visual information relies only on the CLIP visual embedding. Because of this issue, there is a limitation that the image descriptions inherent in the prompt are not sufficiently reflected in the visual embedding space. To tackle this issue, we propose ViPCap, a novel retrieval text-based visual prompt for lightweight image captioning. ViPCap leverages the retrieved text with image information as visual prompts to enhance the ability of the model to capture relevant visual information. By mapping text prompts into the CLIP space and generating multiple randomized Gaussian distributions, our method leverages sampling to explore randomly augmented distributions and effectively retrieves the semantic features that contain image information. These retrieved features are integrated into the image and designated as the visual prompt, leading to performance improvements on the datasets such as COCO, Flickr30k, and NoCaps. Experimental results demonstrate that ViPCap significantly outperforms prior lightweight captioning models in efficiency and effectiveness, demonstrating the potential for a plug-and-play solution.

Recently, linear complexity sequence modeling networks have achieved modeling capabilities similar to Vision Transformers on a variety of computer vision tasks, while using fewer FLOPs and less memory. However, their advantage in terms of actual runtime speed is not significant. To address this issue, we introduce Gated Linear Attention (GLA) for vision, leveraging its superior hardware-awareness and efficiency. We propose direction-wise gating to capture 1D global context through bidirectional modeling and a 2D gating locality injection to adaptively inject 2D local details into 1D global context. Our hardware-aware implementation further merges forward and backward scanning into a single kernel, enhancing parallelism and reducing memory cost and latency. The proposed model, ViG, offers a favorable trade-off in accuracy, parameters, and FLOPs on ImageNet and downstream tasks, outperforming popular Transformer and CNN-based models.

The rapid spread of information in the digital age highlights the critical need for effective fact-checking tools, particularly for languages with limited resources, such as Vietnamese. In response to this challenge, we introduce ViFactCheck, the first publicly available benchmark dataset designed specifically for Vietnamese fact-checking across multiple online news domains. This dataset contains 7,232 human-annotated pairs of claim-evidence combinations sourced from reputable Vietnamese online news, covering 12 diverse topics. It has been subjected to a meticulous annotation process to ensure high quality and reliability, achieving a Fleiss Kappa inter-annotator agreement score of 0.83. Our evaluation leverages state-of-the-art pre-trained and large language models, employing fine-tuning and prompting techniques to assess performance. Notably, the Gemma model demonstrated superior effectiveness, with an impressive macro F1 score of 89.90%, thereby establishing a new standard for fact-checking benchmarks. This result highlights the robust capabilities of Gemma in accurately identifying and verifying facts in Vietnamese. To further promote advances in fact-checking technology and improve the reliability of digital media, we have made the ViFactCheck dataset, model checkpoints, fact-checking pipelines, and source code freely available on GitHub. This initiative aims to inspire further research and enhance the accuracy of information in low-resource languages.

Despite the rapid advancements in text-to-image (T2I) synthesis, enabling precise visual control remains a significant challenge. Existing works attempted to incorporate multi-facet controls (text and sketch), aiming to enhance the creative control over generated images. However, our pilot study reveals that the expressive power of humans far surpasses the capabilities of current methods. Users desire a more versatile approach that can accommodate their diverse creative intents, ranging from controlling individual subjects to manipulating the entire scene composition. We present VersaGen, a generative AI agent that enables versatile visual control in T2I synthesis. VersaGen admits four types of visual controls: i) single visual subject; ii) multiple visual subjects; iii) scene background; iv) any combination of the three above or merely no control at all. We train an adaptor upon a frozen T2I model to accommodate the visual information into the text-dominated diffusion process. We introduce three optimization strategies during the inference phase of VersaGen to improve generation results and enhance user experience. Comprehensive experiments on COCO and Sketchy validate the effectiveness and flexibility of VersaGen, as evidenced by both qualitative and quantitative results.

Text-to-image (T2I) diffusion models have achieved remarkable progress in generating realistic images from textual descriptions. However, ensuring consistent high-quality image generation with complete backgrounds, object appearance, and optimal texture rendering remains challenging. This paper presents a novel fine-grained pixel-level image editing method based on pre-trained diffusion models. The proposed dual-branch architecture, consisting of Guidance and Generation branches, employs U-Net Denoisers and Self-Attention mechanisms. An improved DDIM-like inversion method obtains the latent representation, followed by multiple denoising steps. Cross-branch interactions, such as KV Replacement, Classifier Guidance, and Feature Correspondence, enable precise control while preserving image fidelity. The iterative refinement and reconstruction process facilitates finegrained editing control, supporting attribute modification, image outpainting, style transfer, and face synthesis with Clickand-Drag style editing using masks. Experimental results demonstrate the effectiveness of the proposed approach in enhancing the quality and controllability of T2I-generated images, surpassing existing methods while maintaining attractive computational complexity for practical real-world applications.

Due to the growing complexity of modern Integrated Circuits (ICs), automating hardware design can prevent a significant amount of human error from the engineering process and result in less errors. Verilog is a popular hardware description language for designing and modeling digital systems; thus, Verilog generation is one of the emerging areas of research to facilitate the design process. In this work, we propose VerilogCoder, a system of multiple Artificial Intelligence (AI) agents for Verilog code generation, to autonomously write Verilog code and fix syntax and functional errors using collaborative Verilog tools (i.e., syntax checker, simulator, and waveform tracer). Firstly, we propose a task planner that utilizes a novel Task and Circuit Relation Graph retrieval method to construct a holistic plan based on module descriptions. To debug and fix functional errors, we develop a novel and efficient abstract syntax tree (AST)-based waveform tracing tool, which is integrated within the autonomous Verilog completion flow. The proposed methodology successfully generates 94.2% syntactically and functionally correct Verilog code, surpassing the state-of-the-art methods by 33.9% on the VerilogEval-Human v2 benchmark.

We study a model of temporal voting where there is a fixed time horizon, and at each round the voters report their preferences over the available candidates and a single candidate is selected. Prior work has adapted popular notions of justified representation as well as voting rules that provide strong representation guarantees from the multiwinner election setting to this model. In our work, we focus on the complexity of verifying whether a given outcome offers proportional representation. We show that in the temporal setting verification is strictly harder than in multiwinner voting, but identify natural special cases that enable efficient algorithms.

We develop a method for the efficient verification of neural networks against convolutional perturbations such as blurring or sharpening. To define input perturbations, we use well-known camera shake, box blur and sharpen kernels. We linearly parameterise these kernels in a way that allows for a variation of the perturbation strength while preserving desired kernel properties. To facilitate their use in neural network verification, we develop an efficient way of convolving a given input with the parameterised kernels. The result of this convolution can be used to encode the perturbation in a verification setting by prepending a linear layer to a given network. This leads to tight bounds and a high effectiveness in the resulting verification step. We add further precision by employing input splitting as a branching strategy. We demonstrate that we are able to verify robustness on a number of standard benchmarks where the baseline is unable to provide any safety certificates. To the best of our knowledge, this is the first solution for verifying robustness against specific convolutional perturbations such as camera shake.

Variate tokenization, which independently embeds each variate as separate tokens, has achieved remarkable improvements in multivariate time series forecasting. However, employing self-attention with variate tokens incurs a quadratic computational cost with respect to the number of variates, thus limiting its training efficiency for large-scale applications. To address this issue, we propose VarDrop, a simple yet efficient strategy that reduces the token usage by omitting redundant variate tokens during training. VarDrop adaptively excludes redundant tokens within a given batch, thereby reducing the number of tokens used for dot-product attention while preserving essential information. Specifically, we introduce k-dominant frequency hashing (k-DFH), which utilizes the ranked dominant frequencies in the frequency domain as a hash value to efficiently group variate tokens exhibiting similar periodic behaviors. Then, only representative tokens in each group are sampled through stratified sampling. By performing sparse attention with these selected tokens, the computational cost of scaled dot-product attention is significantly alleviated. Experiments conducted on public benchmark datasets demonstrate that VarDrop outperforms existing efficient baselines.

Video anomaly retrieval (VAR) aims to retrieve pertinent abnormal or normal videos from collections of untrimmed and long videos through cross-modal requires such as textual descriptions and synchronized audios. Cross-modal pre-training (CMP) models, by pre-training on large-scale cross-modal pairs, e.g., image and text, can learn the rich associations between different modalities, and this cross-modal association capability gives CMP an advantage in conventional retrieval tasks. Inspired by this, how to utilize the robust cross-modal association capabilities of CMP in VAR to search crucial visual component from these untrimmed and long videos becomes a critical research problem. Therefore, this paper proposes a VAR method based on CMP models, named VarCMP. First, a unified hierarchical alignment strategy is proposed to constrain the semantic and spatial consistency between video and text, as well as the semantic, temporal, and spatial consistency between video and audio. It fully leverages the efficient cross-modal association capabilities of CMP models by considering cross-modal similarities at multiple granularities, enabling VarCMP to achieve effective all-round information matching for both video-text and video-audio VAR tasks. Moreover, to further solve the problem of untrimmed and long video alignment, an anomaly-biased weighting is devised in the fine-grained alignment, which identifies key segments in untrimmed long videos using anomaly priors, giving them more attention, thereby discarding irrelevant segment information, and achieving more accurate matching with cross-modal queries. Extensive experiments demonstrates high efficacy of VarCMP in both video-text and video-audio VAR tasks, achieving significant improvements on both text-video (UCFCrime-AR) and audio-video (XDViolence-AR) datasets against the best competitors by 5.0% and 5.3% R@1.

With the popularity of 3D volumetric video applications, such as Autonomous Driving, Virtual Reality, and Mixed Reality, current developers have turned to deep learning for compressing volumetric video frames, i.e., point clouds for video upstreaming. The latest deep learning-based solutions offer higher efficiency, lower distortion, and better hardware support compared to traditional ones like MPEG and JPEG. However, privacy threats arise, especially reconstruction attacks targeting to recover the original input point cloud from the intermediate results. In this paper, we design VVRec, to the best of our knowledge, which is the first targeting DL-based Volumetric Video Reconstruction attack scheme. VVRec demonstrates the ability to reconstruct high-quality point clouds from intercepted transmission intermediate results using four well-trained neural network modules we design. Leveraging the latest latent diffusion models with Gamma distribution and a refinement algorithm, VVRec excels in reconstruction quality and color recovery and surpasses existing defenses. We evaluate VVRec using three volumetric video datasets. The results demonstrate that VVRec achieves 64.70dB reconstruction accuracy, with an impressive 46.39% reduction of distortion over baselines.