cs.CV(2024-09-10)

📊 共 15 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

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

🔬 支柱三:空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗
1 gsplat: An Open-Source Library for Gaussian Splatting gsplat:用于高斯溅射的开源库,加速训练并降低内存占用。 gaussian splatting splatting NeRF
2 Neuromorphic spatiotemporal optical flow: Enabling ultrafast visual perception beyond human capabilities 提出神经形态时空光流方法,实现超越人类的超快视觉感知 optical flow spatiotemporal
3 GigaGS: Scaling up Planar-Based 3D Gaussians for Large Scene Surface Reconstruction GigaGS:扩展平面3D高斯到大规模场景表面重建 3D gaussian splatting 3DGS gaussian splatting
4 LEIA: Latent View-invariant Embeddings for Implicit 3D Articulation LEIA:提出一种隐式3D铰接的潜在视角不变嵌入方法,无需运动信息。 NeRF neural radiance field

🔬 支柱二:RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗
5 EyeCLIP: A visual-language foundation model for multi-modal ophthalmic image analysis EyeCLIP:用于多模态眼科图像分析的视觉-语言基础模型 contrastive learning foundation model
6 Loss Distillation via Gradient Matching for Point Cloud Completion with Weighted Chamfer Distance 提出基于梯度匹配的损失蒸馏方法,用于点云补全,并使用加权倒角距离。 distillation scene understanding
7 DetailCLIP: Detail-Oriented CLIP for Fine-Grained Tasks DetailCLIP:面向细节的CLIP模型,提升细粒度分割任务性能 contrastive learning distillation
8 Learning Generative Interactive Environments By Trained Agent Exploration 提出基于强化学习探索的生成交互环境模型GenieRedux,提升视觉保真度和可控性 reinforcement learning world model

🔬 支柱九:具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗
9 LIME: Less Is More for MLLM Evaluation LIME:精简多模态大语言模型评估基准,提升效率与区分度 large language model multimodal
10 Enhancing Long Video Understanding via Hierarchical Event-Based Memory 提出基于分层事件记忆增强的LLM(HEM-LLM)用于提升长视频理解能力 large language model foundation model
11 Shadow Removal Refinement via Material-Consistent Shadow Edges 提出基于材质一致性阴影边缘的阴影去除优化方法 foundation model
12 Aligning Machine and Human Visual Representations across Abstraction Levels 提出一种对齐机器与人类视觉表征的方法,提升模型泛化性和鲁棒性。 foundation model

🔬 支柱四:生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗
13 Human Motion Synthesis_ A Diffusion Approach for Motion Stitching and In-Betweening 提出基于扩散模型的运动缝合与插值方法,生成逼真流畅的人体运动 motion synthesis motion generation multimodal

🔬 支柱一:机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗
14 Test-Time Certifiable Self-Supervision to Bridge the Sim2Real Gap in Event-Based Satellite Pose Estimation 提出基于测试时自监督的事件相机卫星姿态估计方法,弥合Sim2Real差距。 sim2real

🔬 支柱六:视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗
15 When to Extract ReID Features: A Selective Approach for Improved Multiple Object Tracking 提出一种选择性ReID特征提取方法,在多目标跟踪中降低计算开销并提升精度。 feature matching

⬅️ 返回 cs.CV 首页 · 🏠 返回主页