cs.CV（2024-03-27）

📊 共 33 篇论文 | 🔗 15 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (11 🔗4) 支柱九：具身大模型 (Embodied Foundation Models) (9 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗2) 支柱六：视频提取与匹配 (Video Extraction) (3 🔗1) 支柱一：机器人控制 (Robot Control) (2 🔗2) 支柱四：生成式动作 (Generative Motion) (1 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1 🔗1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (11 篇)

#题目一句话要点标签🔗⭐
1 UniDepth: Universal Monocular Metric Depth Estimation 提出UniDepth以解决单目度量深度估计的通用性问题 depth estimation metric depth UniDepth ✅
2 $\mathrm{F^2Depth}$: Self-supervised Indoor Monocular Depth Estimation via Optical Flow Consistency and Feature Map Synthesis 提出F^2Depth以解决室内单目深度估计中的低纹理区域问题 depth estimation monocular depth optical flow
3 ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation 提出ECoDepth以解决单目深度估计中的上下文信息不足问题 depth estimation monocular depth ZoeDepth
4 Modeling uncertainty for Gaussian Splatting 提出随机高斯点云以解决高斯点云的不确定性估计问题 gaussian splatting splatting NeRF
5 SAT-NGP : Unleashing Neural Graphics Primitives for Fast Relightable Transient-Free 3D reconstruction from Satellite Imagery 提出SAT-NGP以解决卫星影像的快速3D重建问题 3D reconstruction NeRF neural radiance field
6 Lift3D: Zero-Shot Lifting of Any 2D Vision Model to 3D 提出Lift3D以实现任意2D视觉模型的3D提升 neural radiance field open-vocabulary open vocabulary
7 SplatFace: Gaussian Splat Face Reconstruction Leveraging an Optimizable Surface 提出SplatFace框架以解决3D人脸重建问题 3D reconstruction gaussian splatting splatting
8 ModaLink: Unifying Modalities for Efficient Image-to-PointCloud Place Recognition 提出ModaLink以解决跨模态图像与点云的场所识别问题 depth estimation ✅
9 SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model 提出SingularTrajectory以解决多任务轨迹预测问题 traversability ✅
10 WALT3D: Generating Realistic Training Data from Time-Lapse Imagery for Reconstructing Dynamic Objects under Occlusion 提出WALT3D以解决城市环境中动态物体遮挡问题 3D reconstruction
11 DVLO: Deep Visual-LiDAR Odometry with Local-to-Global Feature Fusion and Bi-Directional Structure Alignment 提出DVLO以解决视觉与LiDAR数据融合问题 scene flow ✅

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

#题目一句话要点标签🔗⭐
12 LITA: Language Instructed Temporal-Localization Assistant 提出LITA以解决视频中的时间定位问题 large language model multimodal instruction following ✅
13 Toward Interactive Regional Understanding in Vision-Large Language Models 提出RegionVLM以解决视觉语言模型区域理解不足问题 large language model
14 An Evolutionary Network Architecture Search Framework with Adaptive Multimodal Fusion for Hand Gesture Recognition 提出自适应多模态融合的进化网络架构搜索框架以解决手势识别问题 multimodal
15 An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM 提出图像网格方法以解决零-shot视频问答问题 large language model foundation model
16 Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models 提出Mini-Gemini以提升多模态视觉语言模型性能 large language model ✅
17 ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object 提出ImageNet-D以评估神经网络在合成物体上的鲁棒性 foundation model ✅
18 Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding 提出指令对比解码以解决大规模视觉-语言模型中的幻觉问题 multimodal
19 Bringing Textual Prompt to AI-Generated Image Quality Assessment 提出IP-IQA以解决AI生成图像质量评估中的多模态问题 multimodal ✅
20 Addressing Data Annotation Challenges in Multiple Sensors: A Solution for Scania Collected Datasets 提出运动补偿方法以解决多传感器数据标注挑战 TAMP

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
21 Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction 提出Gamba以解决单视图3D重建效率低下问题 Mamba distillation 3D gaussian splatting ✅
22 Scaling Vision-and-Language Navigation With Offline RL 提出基于离线强化学习的视觉语言导航新方法以解决数据稀缺问题 reinforcement learning offline RL offline reinforcement learning
23 Beyond Embeddings: The Promise of Visual Table in Visual Reasoning 提出视觉表以解决视觉推理中的知识缺失问题 representation learning large language model multimodal ✅
24 I2CKD : Intra- and Inter-Class Knowledge Distillation for Semantic Segmentation 提出I2CKD以提升语义分割模型的知识蒸馏效果 teacher-student distillation
25 Duolando: Follower GPT with Off-Policy Reinforcement Learning for Dance Accompaniment 提出Duolando以解决3D舞蹈伴奏生成问题 reinforcement learning
26 LayoutFlow: Flow Matching for Layout Generation 提出LayoutFlow以解决布局生成中的效率问题 flow matching

🔬 支柱六：视频提取与匹配 (Video Extraction) (3 篇)

#题目一句话要点标签🔗⭐
27 EgoNav: Egocentric Scene-aware Human Trajectory Prediction 提出EgoNav以解决人类轨迹预测中的场景感知问题 egocentric egocentric vision human motion
28 Generative Multi-modal Models are Good Class-Incremental Learners 提出生成多模态模型以解决类增量学习中的灾难性遗忘问题 feature matching ✅
29 AIR-HLoc: Adaptive Retrieved Images Selection for Efficient Visual Localisation 提出AIR-HLoc以解决视觉定位中的特征匹配瓶颈问题 feature matching

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
30 Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark 提出Real Acoustic Fields数据集以解决真实环境声学建模问题 sim2real PULSE ✅
31 NeuroPictor: Refining fMRI-to-Image Reconstruction via Multi-individual Pretraining and Multi-level Modulation 提出NeuroPictor以解决fMRI到图像重建的控制不足问题 manipulation ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
32 ParCo: Part-Coordinating Text-to-Motion Synthesis 提出ParCo以解决文本到动作合成中的协调性问题 text-to-motion motion synthesis ✅

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

#题目一句话要点标签🔗⭐
33 NeuSDFusion: A Spatial-Aware Generative Model for 3D Shape Completion, Reconstruction, and Generation 提出NeuSDFusion以解决3D形状生成中的空间一致性问题 spatial relationship ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页