ICCV 2025 Papers

Cancers identified in CT scans are usually accompanied by detailed radiology reports, but publicly available CT datasets often lack these essential reports. This absence limits their usefulness for developing accurate report generation AI. To address this gap, we present AbdomenAtlas 3.0, the first public, high-quality abdominal CT dataset with detailed, expert-reviewed radiology reports. All reports are paired with per-voxel masks and they describe liver, kidney and pancreatic tumors. AbdomenAtlas 3.0 has 9,262 triplets of CT, mask and report--3,955 with tumors. These CT scans come from 17 public datasets. Besides creating the reports for these datasets, we expanded their number of tumor masks by 4.2x, identifying 3,011 new tumor cases. Notably, the reports in AbdomenAtlas 3.0 are more standardized, and generated faster than traditional human-made reports. They provide details like tumor size, location, attenuation and surgical resectability. These reports were created by 12 board-certified radiologists using our proposed RadGPT, a novel framework that converted radiologist-revised tumor segmentation masks into structured and narrative reports. Besides being a dataset creation tool, RadGPT can also become a fully-automatic, segmentation-assisted report generation method. We benchmarked this method and 5 state-of-the-art report generation vision-language models. Our results show that segmentation strongly improves tumor detection in AI-made reports.

While recent online HD mapping methods relieve burdened offline pipelines and solve map freshness, they remain limited by perceptual inaccuracies, occlusion in dense traffic, and an inability to fuse multi-agent observations. We propose RTMap to enhance these single-traversal methods by persistently crowdsourcing a multi-traversal HD map as a self-evolutional memory. On onboard agents, RTMap simultaneously addresses three core challenges in an end-to-end fashion: (1) Uncertainty-aware positional modeling for HD map elements, (2) probabilistic-aware localization w.r.t. the crowdsourced prior-map, and (3) real-time detection for possible road structural changes. Experiments on several public autonomous driving datasets demonstrate our solid performance on both the prior-aided map quality and the localization accuracy, demonstrating our effectiveness of robustly serving downstream prediction and planning modules while gradually improving the accuracy and freshness of the crowdsourced prior-map asynchronously. Our source-code will be made publicly available at https://github.com/CN-ADLab/RTMap.

Remote sensing foundation models largely break away from the traditional paradigm of designing task-specific models, offering greater scalability across multiple tasks. However, they face challenges such as low computational efficiency and limited interpretability, especially when dealing with large-scale remote sensing images. To overcome these, we draw inspiration from heat conduction, a physical process modeling local heat diffusion. Building on this idea, we are the first to explore the potential of using the parallel computing model of heat conduction to simulate the local region correlations in high-resolution remote sensing images, and introduce RS-vHeat, an efficient multi-modal remote sensing foundation model. Specifically, RS-vHeat 1) applies the Heat Conduction Operator (HCO) with a complexity of O(N^ 1.5 ) and a global receptive field, reducing computational overhead while capturing remote sensing object structure information to guide heat diffusion; 2) learns the frequency distribution representations of various scenes through a self-supervised strategy based on frequency domain hierarchical masking and multi-domain reconstruction; 3) significantly improves efficiency and performance over state-of-the-art techniques across 4 tasks and 10 datasets. Compared to attention-based remote sensing foundation models, we reduce memory usage by 84%, FLOPs by 24% and improves throughput by 2.7 times. The code will be made publicly available.

We present ROVI, a high-quality synthetic dataset for instance-grounded text-to-image generation, created by labeling 1M curated web images. Our key innovation is a strategy called re-captioning, focusing on the pre-detection stage, where a VLM (Vision-Language Model) generates comprehensive visual descriptions that are then processed by an LLM (Large Language Model) to extract a flat list of potential categories for OVDs (Open-Vocabulary Detectors) to detect. This approach yields a global prompt inherently linked to instance annotations while capturing secondary visual elements humans typically overlook. Evaluations show that ROVI exceeds existing detection datasets in image quality and resolution while containing two orders of magnitude more categories with an open-vocabulary nature. For demonstrative purposes, a text-to-image model GLIGEN trained on ROVI significantly outperforms state-of-the-art alternatives in instance grounding accuracy, prompt fidelity, and aesthetic quality. Our dataset and reproducible pipeline are available at https://github.com/CihangPeng/ROVI.

Generative image watermarking enables the proactive detection and traceability of generated images. Among existing methods, inversion-based frameworks achieve highly conceal ed watermark embedding by injecting watermarks into the latent representation before the diffusion process. The robustness of this approach hinges on both the embedding mechanism and inversion accuracy. However, prior works have predominantly focused on optimizing the embedding process while overlooking inversion errors, which significantly affect extraction fidelity. In this paper, we address the challenge of inversion errors and propose ROAR, a dual-domain optimization-based framework designed to mitigate errors arising from two key sources: 1) Latent-domain errors, which accumulate across inversion steps due to inherent approximation assumptions. 2) Pixel-domain errors, which result from channel distortions such as JPEG compression. To tackle these issues, we introduce two novel components: A Regeneration-based Optimization (RO) mechanism, which incorporates an optimizable starting latent to minimize latent-domain errors; A Mixture of Experts (MoE)-based distortion-adaptive restoration (AR) network, which effectively recovers watermarked distributions from pixel-level distortions.Extensive experiments demonstrate that ROAR significantly reduces inversion errors and enhances watermark extraction robustness, thereby improving the reliability of generative image watermarking.

Perceiving and autonomously navigating through work zones is a challenging and under-explored problem. Open datasets for this long-tailed scenario are scarce. We propose the ROADWork dataset to learn to recognize, observe, analyze, and drive through work zones. State-of-the-art foundation models fail when applied to work zones. Fine-tuning models on our dataset significantly improves perception and navigation in work zones. With ROADWork, we discover new work zone images with higher precision (+32.5%) at a much higher rate (12.8x) around the world. Open-vocabulary methods fail too, whereas fine-tuned detectors improve performance (+32.2 AP).Vision-Language Models (VLMs) struggle to describe work zones, but fine-tuning substantially improves performance (+36.7 SPICE). Beyond fine-tuning, we show the value of simple techniques. Video label propagation provides additional gains (+2.6 AP) for instance segmentation. While reading work zone signs, composing a detector and text spotter via crop-scaling improves performance (+14.2% 1-NED). Composing work zone detections to provide context further reduces hallucinations (+3.9 SPICE) in VLMs. We predict navigational goals and compute drivable paths from work zone videos. Incorporating road work semantics ensures 53.6% goals have angular error (AE) < 0.5 (+9.9%) and 75.3% pathways have AE < 0.5 (+8.1%).

Visual Language Models (VLMs) have achieved remarkable success in many domains due to their ability to perform step-by-step reasoning. However, progress in the telecommunication (Telecom) domain remains limited, primarily due to the lack of high-quality datasets and domain-specific insights. In this paper, we introduce RMultiplex200K, a multimodal dataset designed to present step-wise reasoning rationales and correctness scores for real-world Telecom questions. This enables VLMs to engage in step-level reasoning and verification using multimodal information, thereby facilitating reliable problem-solving. RMultiplex200K is highly scalable as it is constructed without human annotations, relying instead on our automatic plan-based annotation (ApPA) method, which automatically synthesizes reasoning steps labeled with reward scores. With this dataset, we introduce TC-NAVIGATOR, a new mechanism for training multimodal process reward models to serve as reliable reasoning verifiers for VLMs. For instance, the Qwen-2-VL-72B and Llama-3.2-90B models, which initially achieve only 21.3% and 19.8% respectively on practice Telecom questions, reached 48.5% and 46.1% accuracy, respectively, after training with RMultiplex200K and verifying with TC-NAVIGATOR.

We introduce RIPE, an innovative reinforcement learning-based framework for weakly-supervised training of a keypoint extractor that excels in both detection and description tasks. In contrast to conventional training regimes that depend heavily on artificial transformations, pre-generated models, or 3D data, RIPE requires only a binary label indicating whether paired images represent the same scene.This minimal supervision significantly expands the pool of training data, enabling the creation of a highly generalized and robust keypoint extractor. RIPE utilizes the encoder's intermediate layers for the description of the keypoints with a hyper-column approach to integrate information from different scales. Additionally, we propose a auxiliary loss to enhance the discriminative capability of the learned descriptors.Comprehensive evaluations on standard benchmarks demonstrate that RIPE simplifies data preparation while achieving competitive performance compared to state-of-the-art techniques, marking a significant advancement in robust keypoint extraction and description.Code and data will be made available for research purposes.

The multi-modal 3D semantic occupancy task provides a comprehensive understanding of the scene and has received considerable attention in the field of autonomous driving. However, existing methods mainly focus on processing large-scale voxels, which bring high computational costs and degrade details. Additionally, they struggle to accurately capture occluded targets and distant information. In this paper, we propose a novel LiDAR-Camera 3D semantic occupancy prediction framework called RIOcc, with collaborative feature refinement and multi-scale cross-modal fusion transformer. Specifically, RIOcc encodes multi-modal data into a unified Bird's Eye View (BEV) space, which reduces computational complexity and enhances the efficiency of feature alignment. Then, multi-scale feature processing substantially expands the receptive fields. Meanwhile, in the LiDAR branch, we design the Dual-branch Pooling (DBP) to adaptively enhance geometric features across both the Channel and Grid dimensions. In the camera branch, the Wavelet and Semantic Encoders are developed to extract high-level semantic features with abundant edge and structural information. Finally, to facilitate effective cross-modal complementarity, we develop the Deformable Dual-Attention (DDA) module. Extensive experiments demonstrate that RIOcc achieves state-of-the-art performance, with 54.2 mIoU and 25.9 mIoU on the Occ3D-nuScenes and nuScenes-Occupancy datasets, respectively.

In this paper, we propose RI3D, a novel 3DGS-based approach that harnesses the power of diffusion models to reconstruct high-quality novel views given a sparse set of input images. Our key contribution is separating the view synthesis process into two tasks of reconstructing visible regions and hallucinating missing regions, and introducing two personalized diffusion models, each tailored to one of these tasks. Specifically, one model ('repair') takes a rendered image as input and predicts the corresponding high-quality image, which in turn is used as a pseudo ground truth image to constrain the optimization. The other model ('inpainting') primarily focuses on hallucinating details in unobserved areas. To integrate these models effectively, we introduce a two-stage optimization strategy: the first stage reconstructs visible areas using the repair model, and the second stage reconstructs missing regions with the inpainting model while ensuring coherence through further optimization. Moreover, we augment the optimization with a novel Gaussian initialization method that obtains per-image depth by combining 3D-consistent and smooth depth with highly detailed relative depth. We demonstrate that by separating the process into two tasks and addressing them with the repair and inpainting models, we produce results with detailed textures in both visible and missing regions that outperform state-of-the-art approaches on a diverse set of scenes with extremely sparse inputs.

A single-pass driving clip frequently results in incomplete scanning of the road structure, making reconstructed scene expanding a critical requirement for sensor simulators to effectively regress driving actions. Although contemporary 3D Gaussian Splatting (3DGS) techniques achieve remarkable reconstruction quality, their direct extension through the integration of diffusion priors often introduces cumulative physical inconsistencies and compromises training efficiency. To address these limitations, we present RGE-GS, a novel expansive reconstruction framework that synergizes diffusion-based generation with reward-guided Gaussian integration. The RGE-GS framework incorporates two key innovations: First, we propose a reward network that learns to identify and prioritize consistently generated patterns prior to reconstruction phases, thereby enabling selective retention of diffusion outputs for spatial stability. Second, during the reconstruction process, we devise a differentiated training strategy that automatically adjust Gaussian optimization progress according to scene converge metrics, which achieving better convergence than baseline methods. Extensive evaluations of publicly available datasets demonstrate that RGE-GS achieves state-of-the-art performance in reconstruction quality.

Crowd evacuation simulation is critical for enhancing public safety, and demanded for realistic virtual environments. Current mainstream evacuation models overlook the complex human behaviors that occur during evacuation, such as pedestrian collisions, interpersonal interactions, and variations in behavior influenced by terrain types or individual body shapes. This results in the failure to accurately simulate the escape of people in the real world. In this paper, aligned with the sensory-decision-motor (SDM) flow of the human brain, we propose a real-time 3D crowd evacuation simulation framework that integrates a 3D-adaptive SFM (Social Force Model) Decision Mechanism and a Personalized Gait Control Motor. This framework allows multiple agents to move in parallel and is suitable for various scenarios, with dynamic crowd awareness. Additionally, we introduce Part-level Force Visualization to assist in evacuation analysis. Experimental results demonstrate that our framework supports dynamic trajectory planning and personalized behavior for each agent throughout the evacuation process, and is compatible with uneven terrain. Visually, our method generates evacuation results that are more realistic and plausible, providing enhanced insights for crowd simulation. The code is available at http://cic.tju.edu.cn/faculty/likun/projects/RESCUE.

Traditional image-to-3D models often struggle with scenes containing multipleobjects due to biases and occlusion complexities. To address this challenge, wepresent REPARO, a novel approach for compositional 3D asset generation fromsingle images. REPARO employs a two-step process: first, it extracts individualobjects from the scene and reconstructs their 3D meshes using off-the-shelf imageto-3D models; then, it optimizes the layout of these meshes through differentiablerendering techniques, ensuring coherent scene composition. By integrating optimaltransport-based long-range appearance loss term and high-level semantic loss termin the differentiable rendering, REPARO can effectively recover the layout of 3Dassets. The proposed method can significantly enhance object independence, detailaccuracy, and overall scene coherence. Extensive evaluation of multi-object scenesdemonstrates that our REPARO offers a comprehensive approach to address thecomplexities of multi-object 3D scene generation from single images.

In this paper we tackle a fundamental question: "Can we train latent diffusion models together with the variational auto-encoder (VAE) tokenizer in an end-to-end manner?" Traditional deep-learning wisdom dictates that end-to-end training is often preferable when possible. However, for latent diffusion transformers, it is observed that end-to-end training both VAE and diffusion-model using standard diffusion-loss is ineffective, even causing a degradation in final performance. We show that while diffusion loss is ineffective, end-to-end training can be unlocked through the representation-alignment (REPA) loss -- allowing both VAE and diffusion model to be jointly tuned during the training process. Despite its simplicity, the proposed training recipe (REPA-E) shows remarkable performance; speeding up diffusion model training by over 17x and 45x over REPA and vanilla training recipes, respectively. Interestingly, we observe that end-to-end tuning with REPA-E also improves the VAE itself; leading to improved latent space structure and downstream generation performance. In terms of final performance, our approach sets a new state-of-the-art; achieving FID of 1.26 and 1.83 with and without classifier-free guidance on ImageNet 256x256. Code is available at https://end2end-diffusion.github.io.

We present a novel perspective on learning video embedders for generative modeling: rather than requiring an exact reproduction of an input video, an effective embedder should focus on synthesizing visually plausible reconstructions. This relaxed criterion enables substantial improvements in compression ratios without compromising the quality of downstream generative models. Specifically, we propose replacing the conventional encoder-decoder video embedder with an encoder-generator framework that employs a diffusion transformer (DiT) to synthesize missing details from a compact latent space. Therein, we develop a dedicated latent conditioning module to condition the DiT decoder on the encoded video latent embedding. Our experiments demonstrate that our approach enables superior encoding-decoding performance compared to state-of-the-art methods, particularly as the compression ratio increases. To demonstrate the efficacy of our approach, we report results from our video embedders achieving a temporal compression ratio of up to 32x (8x higher than leading video emebbders) and validate the robustness of this ultra-compact latent space for text-to-video generation, providing a significant efficiency boost in latent diffusion model training and inference.

Commercial video generation models have exhibited realistic, high-fidelity results but are still restricted to limited access.One crucial obstacle for large-scale applications is the expensive training and inference cost.In this paper, we argue that videos contain significantly more redundant information than images, allowing them to be encoded with very few motion latents.Towards this goal, we design an image-conditioned VAE that projects videos into extremely compressed latent space and decode them based on content images. This magic Reducio charm enables 64xreduction of latents compared to a common 2D VAE, without sacrificing the quality.Building upon Reducio-VAE, we can train diffusion models for high-resolution video generation efficiently. Specifically, we adopt a two-stage generation paradigm, first generating a condition image via text-to-image generation, followed by text-image-to-video generation with the proposed Reducio-DiT. Extensive experiments show that our model achieves strong performance in evaluation.More importantly, our method significantly boosts the training and inference efficiency of video LDMs. Reducio-DiT is trained in just 3.2K A100 GPU hours in total and can generate a 16-frame 1024x1024 video clip within 15.5 seconds on a single A100 GPU. Code is available at https://github.com/microsoft/Reducio-VAE.

Radar-camera fusion methods have emerged as a cost-effective approach for 3D object detection but still lag behind LiDAR-based methods in performance. Recent works have focused on employing temporal fusion and Knowledge Distillation (KD) strategies to overcome these limitations. However, existing approaches have not sufficiently accounted for uncertainties arising from object motion or sensor-specific errors inherent in radar and camera modalities. In this work, we propose RCTDistill, a novel cross-modal KD method based on temporal fusion, comprising three key modules: Range-Azimuth Knowledge Distillation (RAKD), Temporal Knowledge Distillation (TKD), and Region-Decoupled Knowledge Distillation (RDKD). RAKD is designed to consider the inherent errors in the range and azimuth directions, enabling effective knowledge transfer from LiDAR features to refine inaccurate BEV representations. TKD mitigates temporal misalignment caused by dynamic objects by aligning historical radar-camera BEV features with LiDAR representations. RDKD enhances feature discrimination by distilling relational knowledge from the teacher model, allowing the student to understand better and differentiate foreground and background features. RCTDistill achieves state-of-the-art radar-camera fusion performance on both the nuScenes and view-of-delft (VoD) datasets, with the fastest inference speed of 26.2 FPS.

Recent research leveraging large-scale pretrained diffusion models has demonstrated the potential of using diffusion features to establish semantic correspondences in images. Inspired by advancements in diffusion-based techniques, we propose a novel zero-shot method for refining point cloud registration algorithms. Our approach leverages correspondences derived from depth images to enhance point feature representations, eliminating the need for a dedicated training dataset. Specifically, we first project the point cloud into depth maps from multiple perspectives and extract implicit knowledge from a pretrained diffusion network as depth diffusion features. These features are then integrated with geometric features obtained from existing methods to establish more accurate correspondences between point clouds. By leveraging these refined correspondences, our approach achieves significantly improved registration accuracy. Extensive experiments demonstrate that our method not only enhances the performance of existing point cloud registration techniques but also exhibits robust generalization capabilities across diverse datasets. Codes are available a https://github.com/zhengcy-lambo/RARE.git.

Self-supervised contrastive learning models, such as CLIP, have set new benchmarks for vision-language models in many downstream tasks. However, their dependency on rigid one-to-one mappings overlooks the complex and often multifaceted relationships between and within texts and images. To this end, we introduce RankCLIP, a novel pretraining method that extends beyond the rigid one-to-one matching framework of CLIP and its variants. By extending the traditional pair-wise loss to list-wise, and leveraging both in-modal and cross-modal ranking consistency, RankCLIP improves the alignment process, enabling it to capture the nuanced many-to-many relationships between and within each modality. Through comprehensive experiments, we demonstrate the effectiveness of RankCLIP in various downstream tasks, notably achieving significant gains in zero-shot classifications over state-of-the-art methods, underscoring the importance of this enhanced learning process.