cs.CV（2026-09-14）

📊 共 33 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (10 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (6 🔗2) 支柱八：物理动画 (Physics-based Animation) (2) 支柱四：生成式动作 (Generative Motion) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Benchmarking Intra-Patient 3D Deformable Multimodal Image Registration 提出全面基准测试以解决多模态图像配准挑战 foundation model multimodal
2 PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models 提出PhysBrain 1.5以解决物理环境理解与预测问题 foundation model multimodal
3 SURE-Map: Self-Correcting Streaming Geometric Foundation Model 提出SURE-Map以解决流式几何重建中的自我修正问题 foundation model ✅
4 V-ICAL Bench: Evaluating Video In-Context Learning for Multimodal Agents in Interactive Environments 提出V-ICAL基准以评估多模态代理在互动环境中的视频上下文学习能力 multimodal
5 Query-Conditioned Spherical Centroid Aggregation for Multimodal Retrieval 提出SCALAR以解决多模态检索中的均匀影响问题 multimodal
6 Hypergraph-Regularized Gramian Volumes for Multimodal Retrieval 提出超图正则化Gramian体积以提升多模态检索性能 multimodal
7 woma: a real-time foundation model and its fine-tuned models for endoscopy 提出woma模型以解决内窥镜实时分析问题 foundation model
8 TwinICL: Diagnosing Multimodal In-Context Learning through Paired Counterfactuals 提出TwinICL以解决多模态上下文学习性能评估问题 multimodal ✅
9 LynnReal-Omni: Native multi-modal Video Generation for Agentic Visual Workflows 提出LynnReal-Omni以解决视频生成控制难题 multimodal instruction following
10 Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning for Circuit Schematic Understanding 提出Circuit-MLLM以解决电路原理图理解问题 large language model multimodal ✅
11 MarKey: Marginal Utility Guided Greedy Keyframe Selection for Long Video Understanding 提出MarKey以解决长视频理解中的关键帧选择问题 large language model multimodal
12 Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation 提出CORAL以解决医疗报告生成中的可解释性问题 large language model multimodal
13 PACE: Progressive Angular-to-Norm Contrastive Embedding 提出PACE框架以解决多模态嵌入中的角度与范数信息优化问题 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
14 Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI 提出解剖基础与泄漏感知的多模态对比学习以解决阿尔茨海默病分类问题 contrastive learning multimodal
15 MambaMPD: A Mamba-Driven Segmentation Framework for Marine Pollution Detection from Remote Sensing Imagery 提出MambaMPD框架以解决海洋污染检测问题 Mamba foundation model
16 VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding 提出VideoScout以解决长视频理解中的证据获取问题 reinforcement learning large language model multimodal
17 AlignUS: MRI-Guided Ultrasound Representation Learning for ALS Classification from Tongue Images 提出AlignUS以解决ALS分类中的MRI依赖问题 representation learning contrastive learning distillation
18 GRAVA: Grounded Reasoning-to-Action Representation and Learning for Autonomous Driving 提出GRAVA框架以解决自主驾驶中的推理与行动问题 reinforcement learning vision-language-action VLA
19 AdaVSkip: Adaptive Visual Token Skipping Across Layers For Efficient MLLMs Inference 提出AdaVSkip以解决多模态大语言模型推理效率问题 reinforcement learning large language model multimodal
20 Long-to-Short Video Evidence Reasoning for Grounded Question Answering 提出LOVER模型以解决视频证据推理中的问答问题 reinforcement learning curriculum learning TAMP
21 Tele360: Real-Time Feed-Forward Human Reconstruction from Sparse Unposed Cameras 提出Tele360以解决实时动态人类重建问题 teacher-student distillation foundation model
22 Reconstructing Is Not Acting: Action-Centric Latent Dynamics Modeling 提出ACT-LAM以解决重建与动作表示不匹配问题 latent dynamics ✅
23 Unsupervised Point Cloud Registration via Training-Time Semantic Guidance 提出CAESAR以解决无监督点云配准中的几何模糊问题 teacher-student distillation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (6 篇)

#题目一句话要点标签🔗⭐
24 SAM3D-Part: Interactive Part Selection and Generation from 3D Objects 提出SAM3D-Part以解决3D对象部分选择与生成问题 sam 3D SAM 3D ✅
25 ViCo-SAM3: Vision-Conditioned Alignment for Open-Vocabulary Camouflaged Object Segmentation 提出ViCo-SAM3以解决开放词汇伪装物体分割中的语义差距问题 open-vocabulary open vocabulary
26 LG-VLN: A Zero-Shot Vision-and-Language Navigation Framework with LangGraph State Orchestration 提出LG-VLN框架以解决连续环境下的视觉语言导航问题 3D reconstruction VLN
27 Multi-Stage NeRF for Efficient 3D Coronary Artery Reconstruction from Two Narrow-Angle Angiographic Projections 提出NeCA++以解决狭角投影下的3D冠状动脉重建问题 3D reconstruction NeRF neural radiance field
28 Robust Multi-Model Fitting through Learning Neighbor Regions 提出LNR框架以解决多模型拟合中的噪声问题 scene reconstruction
29 G-ray: Ray-Level Relative Geometric Position Encoding in Multi-View Vision Transformers under Camera Heterogeneity 提出G-ray以解决多视角视觉变换器中的相对位置编码问题 3D reconstruction ✅

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
30 A 25-$μ$s/inf Event-driven Graph Neural Network Processor with Spatiotemporal Caching and Spline Convolution for Ultra-low-latency AI at the Edge 提出ETHEREAL以解决事件驱动图神经网络加速问题 spatiotemporal
31 VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention 提出VC-Attention以解决低比特注意力的准确性与速度问题 spatiotemporal

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
32 MoVT: Video-Augmented Motion Tokenizer for Text-to-Motion Generation 提出MoVT框架以解决文本驱动3D人类动作生成问题 text-to-motion motion generation motion tokenizer

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
33 What Makes a 3D Scene Editable? A Factorized Benchmark of Fidelity, Locality, Consistency, and Preservation 提出EditBench3D基准以评估3D场景编辑的多维度特性 manipulation 3D gaussian splatting gaussian splatting

⬅️ 返回 cs.CV 首页 · 🏠 返回主页