cs.CV（2024-04-17）

📊 共 24 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (9 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (6 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱四：生成式动作 (Generative Motion) (2 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
1 Text-controlled Motion Mamba: Text-Instructed Temporal Grounding of Human Motion 提出Text-controlled Motion Mamba以解决文本驱动的人类动作定位问题 Mamba motion generation human motion
2 SCL: Towards Domain Generalization via Single-Temporal Multimodal Contrastive Learning for Remote Sensing Change Detection 提出单时域多模态对比学习以解决遥感变化检测问题 contrastive learning foundation model multimodal ✅
3 Multimodal 3D Object Detection on Unseen Domains 提出CLIX$^{3D}$以解决未见领域的多模态3D物体检测问题 contrastive learning scene understanding multimodal
4 When are Foundation Models Effective? Understanding the Suitability for Pixel-Level Classification Using Multispectral Imagery 探讨基础模型在多光谱图像像素级分类中的适用性 masked autoencoder foundation model
5 CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration 提出CU-Mamba以解决图像恢复中的长距离依赖和计算成本问题 Mamba SSM state space model
6 Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives 提出对比学习方法以解决复合图像检索中的正负样本不足问题 dreamer contrastive learning large language model ✅
7 MaeFuse: Transferring Omni Features with Pretrained Masked Autoencoders for Infrared and Visible Image Fusion via Guided Training 提出MaeFuse以解决红外与可见光图像融合问题 masked autoencoder MAE
8 A Progressive Framework of Vision-language Knowledge Distillation and Alignment for Multilingual Scene 提出DC-CLIP以解决多语言视觉-语言模型的高延迟与内存占用问题 distillation
9 Equivariant Spatio-Temporal Self-Supervision for LiDAR Object Detection 提出时空等变自监督学习框架以提升LiDAR目标检测性能 representation learning scene flow

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

#题目一句话要点标签🔗⭐
10 RainyScape: Unsupervised Rainy Scene Reconstruction using Decoupled Neural Rendering 提出RainyScape以解决多视角雨天场景重建问题 3D gaussian splatting gaussian splatting splatting
11 DeblurGS: Gaussian Splatting for Camera Motion Blur 提出DeblurGS以解决运动模糊图像的3D重建问题 3D gaussian splatting gaussian splatting splatting
12 Predicting Long-horizon Futures by Conditioning on Geometry and Time 通过几何和时间条件预测长时间未来传感器观测 monocular depth TAMP
13 SLAIM: Robust Dense Neural SLAM for Online Tracking and Mapping 提出SLAIM以解决NeRF-SLAM跟踪性能不足问题 NeRF neural radiance field
14 REACTO: Reconstructing Articulated Objects from a Single Video 提出Quasi-Rigid Blend Skinning以解决单视频重建关节物体问题 3D reconstruction neural radiance field ✅
15 A Subspace-Constrained Tyler's Estimator and its Applications to Structure from Motion 提出子空间约束的Tyler估计器以解决运动结构问题 3D reconstruction

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
16 Exploring the Transferability of Visual Prompting for Multimodal Large Language Models 提出可转移视觉提示以提升多模态大语言模型性能 large language model multimodal
17 Rethinking 3D Dense Caption and Visual Grounding in A Unified Framework through Prompt-based Localization 提出统一框架3DGCTR以解决3D视觉定位与密集描述问题 visual grounding ✅
18 Fact :Teaching MLLMs with Faithful, Concise and Transferable Rationales 提出Fact以解决多模态大语言模型的推理透明性问题 large language model multimodal

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#题目一句话要点标签🔗⭐
19 LADDER: An Efficient Framework for Video Frame Interpolation 提出LADDER框架以高效解决视频帧插值问题 motion generation
20 Closely Interactive Human Reconstruction with Proxemics and Physics-Guided Adaption 提出基于亲密交互和物理引导的适应方法以解决人类重建问题 penetration VQ-VAE ✅

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
21 NeuroHash: A Hyperdimensional Neuro-Symbolic Framework for Spatially-Aware Image Hashing and Retrieval 提出NeuroHash以解决图像检索中的空间关系问题 manipulation spatial relationship
22 VBR: A Vision Benchmark in Rome 提出VBR基准数据集以推动视觉里程计和SLAM研究 quadruped visual odometry

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
23 State-space Decomposition Model for Video Prediction Considering Long-term Motion Trend 提出状态空间分解模型以解决视频预测中的长期运动趋势问题 motion prediction

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
24 SPAMming Labels: Efficient Annotations for the Trackers of Tomorrow 提出SPAM以解决视频轨迹标注效率低下问题 spatiotemporal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页