cs.CV（2026-10-16）

📊 共 34 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

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) (9 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (9 🔗4) 支柱八：物理动画 (Physics-based Animation) (2) 支柱一：机器人控制 (Robot Control) (2) 支柱四：生成式动作 (Generative Motion) (2 🔗1)

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

#题目一句话要点标签🔗⭐
1 Persistent Planar Memory for Video World Models 提出PPMem以解决视频世界模型的持久3D记忆问题 world model world models metric depth ✅
2 A100: A Hybrid Vision Foundation Model for Agricultural Plant Image Analysis 提出A100以解决农业植物图像分析中的数据不足问题 distillation foundation model
3 VTrajEmbed: Aligning Text with Video Object Trajectories via Spatiotemporal Embeddings 提出VTrajEmbed以解决文本与视频对象轨迹对齐问题 contrastive learning spatiotemporal multimodal
4 Execution Before Memory: Diagnosing Memory Failures in Video World Models 提出MIST与M3-Agent以解决视频世界模型中的记忆失败问题 world model world models
5 Momentum: Persistent Dynamics for Video World Models 提出Momentum以解决视频世界模型动态场景一致性问题 world model world models
6 LithoWorld: An Endpoint World Model for Lithography Process-Window Recovery 提出LithoWorld以解决光刻工艺窗口恢复问题 world model world models
7 Synthetic Image Oversampling for Rare-Class Representation Learning in Deep Facial Phenotyping 提出合成图像过采样以解决稀有类表征学习问题 representation learning multimodal
8 COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning 提出COMET以解决任务无关的持续视觉-语言学习问题 distillation multimodal
9 Stabilizing Illumination Conditioning in SSM-Based Low-Light Image Enhancement with Bottleneck Prompt Sidecars 提出瓶颈提示侧车以解决SSM基础低光图像增强中的照明条件问题 Mamba SSM
10 Reliability-Consistent Appearance Modeling for RGB-Infrared Multi-Object Tracking 提出RCA-MOT框架以解决RGB-红外多目标跟踪中的不可靠性问题 representation learning multimodal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (9 篇)

#题目一句话要点标签🔗⭐
11 Frame-Consistent Feed-Forward 3D Gaussian Splatting from Sparse Unposed Images 提出框架一致的前馈3D高斯点云重建以解决稀疏图像问题 3D gaussian splatting gaussian splatting splatting
12 X-Splat: Unified Gaussian Splatting for Joint RGB and X-ray 3D Reconstruction 提出统一高斯点云以解决RGB与X射线3D重建问题 3D reconstruction gaussian splatting splatting
13 VISTA: A Multimodal Egocentric Benchmark for Goal-Oriented Perception and Assistance to Blind and Low-Vision Users 提出VISTA基准以解决盲人和低视力用户的目标导向感知问题 scene understanding egocentric multimodal
14 Look, Reason, Draw a Region: Recursive Refinement for Spatial Context Prediction 提出Spatial-TRM以解决空间上下文预测问题 open-vocabulary open vocabulary spatial relationship
15 Direct Conditional Parameterization for N-Dimensional Splatting 提出直接条件参数化以提升N维Splatting性能 3D gaussian splatting gaussian splatting splatting ✅
16 ReVision: Reprogramming Image Generators for Dense Visual Perception 提出ReVision以解决多任务视觉感知中的交叉干扰问题 depth estimation multimodal
17 Video Diffusion Avatars: Taming Video Diffusion Models into Controllable 3D Avatars 提出视频扩散化身以解决单图像生成3D头像问题 3DGS physically plausible
18 Learning to Discriminate Scene Structures Makes Self-Supervised Depth Learning Scalable 提出MaSS框架以解决自监督深度学习的可扩展性问题 depth estimation monocular depth
19 LonGaussian: Tracking Foundation Priors for Long-Context Feedforward Gaussian Reconstruction 提出LonGaussian以解决长视频3D重建问题 3D reconstruction

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
20 ANCHOR: Audio-Visually Grounded Chain-of-Thought Reasoning Benchmark 提出ANCHOR基准以解决多模态推理中的证据对齐问题 large language model multimodal chain-of-thought
21 GeoReason: Bridging Logical Reasoning and Spatial Fidelity in Remote Sensing Segmentation 提出GeoReason以解决遥感分割中的逻辑推理与空间保真问题 large language model multimodal visual grounding ✅
22 ADA 2025: A Multimodal Dataset for Water-Deficit Stress Detection in Sugar Beets 发布ADA 2025数据集以解决水分缺乏应激检测问题 multimodal
23 Enhancing RAG-based Medical VQA with Gated Multimodal Representations 提出多模态RAG框架以提升医学视觉问答的准确性 multimodal
24 Xiaomi-Lens-0 Technical Report 提出Xiaomi-Lens-0以解决多模态语言模型区域理解不足问题 large language model multimodal chain-of-thought ✅
25 MLLM Frame Subset Ensembling for Audio-Visual Video QA and MLLM-based Reranking for Ad-hoc Video Search in TRECVID 2025 提出基于MLLM的框架以提升视频问答和检索性能 foundation model multimodal
26 From GAN-Based Models to GPT-5: A Taxonomy-Guided Study of Human Performance in Detecting Deepfake Face Images 提出深度伪造面孔伪影分类法以提升检测效果 large language model multimodal
27 CatchUp: Depth-Aware Re-Admission of Visual Tokens for Efficient Large Vision-Language Models 提出CatchUp以解决大规模视觉语言模型的视觉令牌效率问题 multimodal ✅
28 BRICE: Brain-inspired Recurrent Illusory Contour Encoder for Point Cloud Completion 提出BRICE以解决点云补全中的信息缺失问题 foundation model ✅

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

#题目一句话要点标签🔗⭐
29 STC-MGNet: Spatial-Temporal-Channel Attention-Enhanced Multimodal Graph Network for Pedestrian Intention Prediction 提出STC-MGNet以解决行人意图预测中的时空信息捕捉问题 spatiotemporal multimodal
30 STRIDE: Evaluating Spatiotemporal Reasoning in Driving Edge Cases 提出STRIDE以解决自动驾驶中的时空推理问题 spatiotemporal

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
31 Diversity Without Labels: Task-Agnostic Coreset Selection Across Perception and Action 提出CoreDNA以解决无标签数据选择问题 manipulation mobile manipulation imitation learning
32 I2V-DETACH: Source Grounding Detachment for Unauthorized Image-to-Video Generation 提出I2V-DETACH以解决图像到视频生成中的未授权操控问题 manipulation

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#题目一句话要点标签🔗⭐
33 HAPACT: A Benchmark For Human-Centric Physical Impact Localization in Movies 提出HAPACT基准以解决人类中心物理冲击定位问题 contact-aware SMPL SMPL-X
34 Agentic Physics-Grounded Video Generation via Memory-Augmented In-Context Reasoning 提出PhysRepair框架以解决视频生成中的物理不一致性问题 physically plausible spatiotemporal ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页