cs.CV（2024-03-26）

📊 共 39 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (11 🔗4) 支柱三：空间感知与语义 (Perception & Semantics) (10 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (5) 支柱六：视频提取与匹配 (Video Extraction) (4 🔗2) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (2 🔗1) 支柱八：物理动画 (Physics-based Animation) (2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
1 Leveraging Near-Field Lighting for Monocular Depth Estimation from Endoscopy Videos 利用近场照明提升内窥镜视频的单目深度估计 teacher-student depth estimation monocular depth
2 Rotate to Scan: UNet-like Mamba with Triplet SSM Module for Medical Image Segmentation 提出Triplet Mamba-UNet以解决医学图像分割中的特征提取不足问题 Mamba SSM state space model
3 OCAI: Improving Optical Flow Estimation by Occlusion and Consistency Aware Interpolation 提出OCAI以解决光流估计中的遮挡与一致性问题 teacher-student optical flow
4 Heracles: A Hybrid SSM-Transformer Model for High-Resolution Image and Time-Series Analysis 提出Heracles模型以解决高分辨率图像和时间序列分析问题 Mamba SSM state space model ✅
5 PlainMamba: Improving Non-Hierarchical Mamba in Visual Recognition 提出PlainMamba以提升视觉识别中的非层次化模型性能 Mamba SSM state space model ✅
6 Serpent: Scalable and Efficient Image Restoration via Multi-scale Structured State Space Models 提出Serpent以解决高分辨率图像恢复效率问题 SSM state space model
7 Superior and Pragmatic Talking Face Generation with Teacher-Student Framework 提出SuperFace框架以解决复杂输入下的生成效率与质量问题 teacher-student distillation
8 Efficient Image Pre-Training with Siamese Cropped Masked Autoencoders 提出CropMAE以解决视频数据依赖问题 masked autoencoder MAE ✅
9 Integrating Mamba Sequence Model and Hierarchical Upsampling Network for Accurate Semantic Segmentation of Multiple Sclerosis Legion 提出Mamba HUNet以解决多发性硬化病灶的语义分割问题 Mamba state space model
10 ReMamber: Referring Image Segmentation with Mamba Twister 提出ReMamber以解决多模态图像分割中的长距离依赖问题 Mamba ✅
11 Neural Clustering based Visual Representation Learning 提出基于神经聚类的特征提取方法以改善视觉表示学习 representation learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (10 篇)

#题目一句话要点标签🔗⭐
12 Move as You Say, Interact as You Can: Language-guided Human Motion Generation with Scene Affordance 提出基于场景可供性的人类动作生成框架以解决多模态信号挑战 affordance motion diffusion model motion diffusion
13 DN-Splatter: Depth and Normal Priors for Gaussian Splatting and Meshing 提出深度与法线先验以提升高保真3D重建效果 depth estimation 3D gaussian splatting 3D reconstruction
14 2D Gaussian Splatting for Geometrically Accurate Radiance Fields 提出2D高斯点云以解决3D高斯表面重建不准确问题 3D gaussian splatting 3DGS gaussian splatting
15 OVER-NAV: Elevating Iterative Vision-and-Language Navigation with Open-Vocabulary Detection and StructurEd Representation 提出OVER-NAV以解决迭代视觉与语言导航中的记忆利用问题 open-vocabulary open vocabulary VLN
16 NeRF-HuGS: Improved Neural Radiance Fields in Non-static Scenes Using Heuristics-Guided Segmentation 提出HuGS以解决NeRF在动态场景中的干扰问题 NeRF neural radiance field scene reconstruction ✅
17 Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving 提出3D$^2$Fool以解决单目深度估计的物理对抗攻击问题 depth estimation monocular depth
18 EgoLifter: Open-world 3D Segmentation for Egocentric Perception 提出EgoLifter以解决自我中心感知中的3D场景分割问题 3D reconstruction egocentric
19 Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians 提出Octree-GS以解决3D高斯渲染中的一致性问题 3D gaussian splatting gaussian splatting splatting
20 Track Everything Everywhere Fast and Robustly 提出CaDeX++以解决视频跟踪效率与鲁棒性问题 depth estimation monocular depth foundation model
21 DeepMIF: Deep Monotonic Implicit Fields for Large-Scale LiDAR 3D Mapping 提出深度单调隐式场以解决稀疏LiDAR数据的3D映射问题 implicit representation

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
22 Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography 提出CT-RATE数据集与CT-CLIP框架以解决3D医学影像AI的挑战 large language model foundation model multimodal
23 Assessment of Multimodal Large Language Models in Alignment with Human Values 提出Ch3Ef评估策略以解决多模态大语言模型与人类价值观对齐问题 large language model multimodal
24 Multimodal CLIP Inference for Meta-Few-Shot Image Classification 提出多模态CLIP推理以提升元少样本图像分类性能 foundation model multimodal
25 The Solution for the CVPR 2023 1st foundation model challenge-Track2 提出跨模态交通检索解决方案以应对图像域间问题 foundation model
26 Hierarchical Light Transformer Ensembles for Multimodal Trajectory Forecasting 提出层次化轻量变换器集成以解决多模态轨迹预测问题 multimodal

🔬 支柱六：视频提取与匹配 (Video Extraction) (4 篇)

#题目一句话要点标签🔗⭐
27 MMVP: A Multimodal MoCap Dataset with Vision and Pressure Sensors 提出MMVP数据集以解决快速人类运动捕捉中的足部接触识别问题 SMPL human motion multimodal ✅
28 EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimation 提出EgoPoseFormer以解决立体自视角3D人体姿态估计问题 egocentric ✅
29 A Survey on 3D Egocentric Human Pose Estimation 综述3D自我中心人类姿态估计的研究现状与挑战 egocentric
30 A Personalized Video-Based Hand Taxonomy: Application for Individuals with Spinal Cord Injury 提出个性化视频基础手部分类法以解决脊髓损伤者的手功能评估问题 egocentric

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
31 DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions 提出DiffH2O以合成自然手-物体交互 manipulation physically plausible
32 Fake or JPEG? Revealing Common Biases in Generated Image Detection Datasets 揭示生成图像检测数据集中的偏差以提高检测性能 manipulation
33 Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge 提出一种新方法以无先验知识定位声音源 OSC ✅

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
34 Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency 提出ReVoRF框架以解决伪新视图合成中的不可靠性问题 geometric consistency ✅
35 ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis 提出ConvoFusion以解决共语手势合成的语义对齐问题 human motion

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

#题目一句话要点标签🔗⭐
36 Towards 3D Vision with Low-Cost Single-Photon Cameras 提出基于低成本单光子相机的3D形状重建方法 PULSE
37 Random-coupled Neural Network 提出随机耦合神经网络以解决脉冲耦合神经网络的计算效率问题 PULSE

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
38 SpectralWaste Dataset: Multimodal Data for Waste Sorting Automation 提出SpectralWaste数据集以解决废物分类自动化问题 HSI multimodal

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
39 InterHandGen: Two-Hand Interaction Generation via Cascaded Reverse Diffusion 提出InterHandGen以解决双手交互生成问题 classifier-free guidance penetration hand reconstruction

⬅️ 返回 cs.CV 首页 · 🏠 返回主页