I. Human Subject Resources

Datasets and benchmarks for perceiving, identifying, and geometrically modeling humans as observable subjects.

## Browse Categories

I.1 Visual Human Observation	I.2 Multisensory Human Sensing
I.3 Human Identity Understanding	I.4 Renderable Human Geometry

--- ## I.1 Visual Human Observation *Visible human appearance, attributes, localized perception, and language-facing visual understanding.* | Resource | Type | Venue | Paper | Paper Page | Website | |---|:---:|:---:|---|:---:|:---:| | MHPR | Benchmark | arXiv 2026 | MHPR: Multidimensional Human Perception and Reasoning Benchmark for Large Vision-Languate Models | [:page_facing_up:](https://arxiv.org/abs/2605.03485 "Paper page") | - | | Face-Human-Bench | Benchmark | NeurIPS 2025 | Face-Human-Bench: A Comprehensive Benchmark of Face and Human Understanding for Multi-Modal Assistants | [:page_facing_up:](https://doi.org/10.52202/085713-0322 "Paper page") | [:octocat:](https://github.com/lxq1000/Face-Human-Bench "GitHub") [:house:](https://face-human-bench.github.io/ "Homepage") | | FaceBench | Dataset + Benchmark | CVPR 2025 | FaceBench: A Multi-View Multi-Level Facial Attribute VQA Dataset for Benchmarking Face Perception MLLMs | [:page_facing_up:](https://doi.org/10.1109/cvpr52734.2025.00855 "Paper page") | [:octocat:](https://github.com/CVI-SZU/FaceBench "GitHub") | | HaGRID | Dataset | WACV 2024 | HaGRID-HAnd Gesture Recognition Image Dataset | [:page_facing_up:](https://doi.org/10.1109/wacv57701.2024.00451 "Paper page") | [:octocat:](https://github.com/hukenovs/hagrid "GitHub") | | CelebV-Text | Dataset | CVPR 2023 | CelebV-Text: A Large-Scale Facial Text-Video Dataset | [:page_facing_up:](https://openaccess.thecvf.com/content/CVPR2023/html/Yu_CelebV-Text_A_Large-Scale_Facial_Text-Video_Dataset_CVPR_2023_paper.html "Paper page") | [:house:](https://celebv-text.github.io/ "Homepage") | | StyleGAN-Human | Dataset | ECCV 2022 | StyleGAN-Human: A Data-Centric Odyssey of Human Generation | [:page_facing_up:](https://arxiv.org/abs/2204.11823 "Paper page") | [:octocat:](https://github.com/stylegan-human/StyleGAN-Human "GitHub") | | CelebV-HQ | Dataset | ECCV 2022 | CelebV-HQ: A Large-Scale Video Facial Attributes Dataset | [:page_facing_up:](https://doi.org/10.1007/978-3-031-20071-7_38 "Paper page") | [:house:](https://celebv-hq.github.io/ "Homepage") |

--- ## I.2 Multisensory Human Sensing *Complementary sensing signals for recovering human state beyond camera-only observation.* | Resource | Type | Venue | Paper | Paper Page | Website | |---|:---:|:---:|---|:---:|:---:| | M4Human | Benchmark | CVPR 2026 | M4Human: A Large-Scale Multimodal mmWave Radar Benchmark for Human Mesh Reconstruction | [:page_facing_up:](https://doi.org/10.48550/arxiv.2512.12378 "Paper page") | [:octocat:](https://github.com/FanJunqiao/M4Human "GitHub") [:house:](https://fanjunqiao.github.io/M4Human-site/ "Homepage") | | MMVR | Dataset + Benchmark | ECCV 2024 | MMVR: Millimeter-wave Multi-View Radar Dataset and Benchmark for Indoor Perception | [:page_facing_up:](https://doi.org/10.1007/978-3-031-72986-7_18 "Paper page") | [:house:](https://zenodo.org/records/12611978 "Homepage") | | RELI11D | Dataset | CVPR 2024 | RELI11D: A Comprehensive Multimodal Human Motion Dataset and Method | [:page_facing_up:](https://doi.org/10.1109/cvpr52733.2024.00219 "Paper page") | [:octocat:](https://github.com/yanmn/RELI11DandLIER "GitHub") [:house:](http://www.lidarhumanmotion.net/reli11d/ "Homepage") | | RT-Pose | Benchmark | ECCV 2024 | RT-Pose: A 4D Radar Tensor-based 3D Human Pose Estimation and Localization Benchmark | [:page_facing_up:](https://doi.org/10.1007/978-3-031-73036-8_7 "Paper page") | [:octocat:](https://github.com/ipl-uw/RT-POSE "GitHub") | | MM-Fi | Dataset | NeurIPS 2023 | MM-Fi: Multi-Modal Non-Intrusive 4D Human Dataset for Versatile Wireless Sensing | [:page_facing_up:](https://doi.org/10.48550/arxiv.2305.10345 "Paper page") | [:octocat:](https://github.com/ybhbingo/MMFi_dataset "GitHub") | | HuPR | Benchmark | WACV 2023 | Hupr: A benchmark for human pose estimation using millimeter wave radar | [:page_facing_up:](https://doi.org/10.1109/wacv56688.2023.00567 "Paper page") | [:octocat:](https://github.com/robert80203/HuPR-A-Benchmark-for-Human-Pose-Estimation-Using-Millimeter-Wave-Radar "GitHub") | | mRI | Dataset | NeurIPS 2022 | mRI: Multi-modal 3D Human Pose Estimation Dataset using mmWave, RGB-D, and Inertial Sensors | [:page_facing_up:](https://doi.org/10.48550/arxiv.2210.08394 "Paper page") | [:octocat:](https://github.com/SizheAn/mRI "GitHub") | | HuMMan | Dataset | ECCV 2022 | HuMMan: Multi-Modal 4D Human Dataset for Versatile Sensing and Modeling | [:page_facing_up:](https://doi.org/10.1007/978-3-031-20071-7_33 "Paper page") | [:octocat:](https://github.com/MotrixLab/humman_toolbox "GitHub") [:house:](https://caizhongang.com/projects/HuMMan/ "Homepage") | | mmBody | Dataset + Benchmark | ACM MM 2022 | mmbody benchmark: 3d body reconstruction dataset and analysis for millimeter wave radar | [:page_facing_up:](https://doi.org/10.1145/3503161.3548262 "Paper page") | [:octocat:](https://github.com/Chen3110/mmBody "GitHub") [:house:](https://chen3110.github.io/mmbody/ "Homepage") |

--- ## I.3 Human Identity Understanding *Person-level matching and retrieval across conditions, viewpoints, sensors, and queries.* | Resource | Type | Venue | Paper | Paper Page | Website | |---|:---:|:---:|---|:---:|:---:| | AG-VPReID | Benchmark | CVPR 2025 | AG-VPReID: A Challenging Large-Scale Benchmark for Aerial-Ground Video-based Person Re-Identification | [:page_facing_up:](https://doi.org/10.1109/cvpr52734.2025.00124 "Paper page") | [:octocat:](https://github.com/agvpreid25/AG-VPReID-Net "GitHub") [:house:](https://agvpreid25.github.io/ "Homepage") | | L2RW | Benchmark | CVPR 2025 | From Laboratory to Real World: A New Benchmark Towards Privacy-Preserved Visible-Infrared Person Re-Identification | [:page_facing_up:](https://doi.org/10.1109/cvpr52734.2025.00825 "Paper page") | [:octocat:](https://github.com/Joey623/L2RW "GitHub") | | MP-ReID | Benchmark | ICCV 2025 | Multi-modal Multi-platform Person Re-Identification: Benchmark and Method | [:page_facing_up:](https://doi.org/10.1109/iccv51701.2025.00955 "Paper page") | [:house:](https://mp-reid.github.io/ "Homepage") | | LLCM | Benchmark | CVPR 2023 | Diverse Embedding Expansion Network and Low-Light Cross-Modality Benchmark for Visible-Infrared Person Re-identification | [:page_facing_up:](https://doi.org/10.1109/cvpr52729.2023.00214 "Paper page") | [:octocat:](https://github.com/ZYK100/LLCM "GitHub") | | MALS | Benchmark | ACM MM 2023 | Towards Unified Text-based Person Retrieval: A Large-Scale Multi-Attribute and Language Search Benchmark | [:page_facing_up:](https://doi.org/10.1145/3581783.3611709 "Paper page") | [:octocat:](https://github.com/Shuyu-XJTU/APTM "GitHub") |

--- ## I.4 Renderable Human Geometry *Three-dimensional and radiance-aware human assets for reconstruction, animation, and rendering.* | Resource | Type | Venue | Paper | Paper Page | Website | |---|:---:|:---:|---|:---:|:---:| | VolHuMe | Dataset + Benchmark | arXiv 2026 | VolHuMe: a High-Resolution Large Scale Dataset of Volumetric Human Meshes | [:page_facing_up:](https://arxiv.org/abs/2606.23062 "Paper page") | [:house:](https://giuli13.github.io/volhume-website/ "Homepage") | | MVHumanNet++ | Dataset | TPAMI 2026 | MVHumanNet++: A Large-scale Dataset of Multi-view Daily Dressing Human Captures with Richer Annotations for 3D Human Digitization | [:page_facing_up:](https://arxiv.org/abs/2505.01838 "Paper page") | [:house:](https://kevinlee09.github.io/research/MVHumanNet++/ "Homepage") | | HumanOLAT | Dataset + Benchmark | ICCV 2025 | HumanOLAT: A Large-Scale Dataset for Full-Body Human Relighting and Novel-View Synthesis | [:page_facing_up:](https://openaccess.thecvf.com/content/ICCV2025/html/Teufel_HumanOLAT_A_Large-Scale_Dataset_for_Full-Body_Human_Relighting_and_Novel-View_ICCV_2025_paper.html "Paper page") | [:house:](https://vcai.mpi-inf.mpg.de/projects/HumanOLAT/ "Homepage") | | HuGe100K | Dataset | CVPR 2025 | IDOL: Instant Photorealistic 3D Human Creation from a Single Image | [:page_facing_up:](https://openaccess.thecvf.com/content/CVPR2025/html/Zhuang_IDOL_Instant_Photorealistic_3D_Human_Creation_from_a_Single_Image_CVPR_2025_paper.html "Paper page") | [:octocat:](https://github.com/yiyuzhuang/IDOL "GitHub") | | WildAvatar | Dataset + Benchmark | CVPR 2025 | WildAvatar: Learning In-the-wild 3D Avatars from the Web | [:page_facing_up:](https://openaccess.thecvf.com/content/CVPR2025/html/Huang_WildAvatar_Learning_In-the-wild_3D_Avatars_from_the_Web_CVPR_2025_paper.html "Paper page") | [:house:](https://wildavatar.github.io/ "Homepage") | | PKU-DyMVHumans | Dataset + Benchmark | CVPR 2024 | PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling | [:page_facing_up:](https://arxiv.org/abs/2403.16080 "Paper page") | [:house:](https://pku-dymvhumans.github.io/ "Homepage") | | Codec Avatar Studio | Dataset | NeurIPS 2024 | Codec Avatar Studio: Paired Human Captures for Complete, Driveable, and Generalizable Avatars | [:page_facing_up:](https://proceedings.neurips.cc/paper_files/paper/2024/hash/9712b78386cebdc3db7f1a48c2d20edb-Abstract-Datasets_and_Benchmarks_Track.html "Paper page") | [:octocat:](https://github.com/facebookresearch/goliath "GitHub") | | DNA-Rendering | Dataset + Benchmark | ICCV 2023 | DNA-Rendering: A Diverse Neural Actor Repository for High-Fidelity Human-centric Rendering | [:page_facing_up:](https://arxiv.org/abs/2307.10173 "Paper page") | [:house:](https://dna-rendering.github.io/ "Homepage") | | SynBody | Dataset | ICCV 2023 | SynBody: Synthetic Dataset with Layered Human Models for 3D Human Perception and Modeling | [:page_facing_up:](https://arxiv.org/abs/2303.17368 "Paper page") | [:house:](https://synbody.github.io/SynBody/ "Homepage") [🤗](https://huggingface.co/datasets/caizhongang/SynBody "Hugging Face") | | HumanRF | Dataset | TOG 2023 | HumanRF: High-Fidelity Neural Radiance Fields for Humans in Motion | [:page_facing_up:](https://doi.org/10.1145/3592415 "Paper page") | [:house:](https://actors-hq.com/ "Homepage") | | NeRSemble | Dataset + Benchmark | TOG 2023 | NeRSemble: Multi-view Radiance Field Reconstruction of Human Heads | [:page_facing_up:](https://doi.org/10.1145/3592455 "Paper page") | [:house:](https://tobias-kirschstein.github.io/nersemble/ "Homepage") |

---

All Survey Resources · Human Dynamics Resources →