Awesome

Human Body Reconstruction

There are tremendous amount of papers on human body digitization in recent years. Below shows my collection of papers organized in a reverse-chronological order.

Contents

1. Multi-view Reconstruction
   • Deep Learning
   • Shape from Silhouette
   • Multi-view Stereo
   • Photometric Stereo
   • Template based Approaches
   • Parametric Models
2. Single-view Reconstruction
3. 4D Scans
4. Dataset and Code

Multi-view Reconstruction

1. Deep Learning

<b>[ICCV19] Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Images</b> [pdf]

<b>[ICCV19] TexturePose: Supervising Human Mesh Estimation with Texture Consistency</b> [pdf]

<b>[ICCV19] Human Mesh Recovery From Monocular Images via a Skeleton-Disentangled Representation</b> [pdf]

<b>[ECCV18] Deep Volumetric Video From Very Sparse Multi-View Performance Capture</b> [pdf]

<b>[ECCV18] Bodynet: Volumetric inference of 3d human body shapes</b> [pdf]

<b>[ECCV18] Volumetric performance capture from minimal camera viewpoints</b> [pdf]

2. Shape from Silhouette

<b>[CGF16] 3d body shapes estimation from dressed-human silhouettes</b> [pdf]

<b>[ICCV15] Interactive visual hull refinement for specular and transparent object surface reconstruction</b> [pdf]

<b>[HPG13] Real-time high-resolution sparse voxelization with application to image-based modeling</b> [pdf]

<b>[TOG08] Articulated Mesh Animation from Multi-view Silhouettes</b> [project page] [pdf] [data]

<b>[ECCV06] Carved visual hulls for image-based modeling</b> [pdf]

<b>[3DPVT06] Visual shapes of silhouette sets</b> [pdf]

<b>[CVIU04] Silhouette and stereo fusion for 3d object modeling</b> [pdf]

<b>[CVPR03] Visual hull alignment and refinement across time: A 3d reconstruction algorithm combining shape-from-silhouette with stereo</b> [pdf]

<b>[CGIT00] Image-based visual hulls</b> [project page][pdf]

3. Multi-view Stereo

<b>[UIST16] Holoportation: Virtual 3d teleportation in real-time</b> [project page] [pdf]

<b>[SIGGRAPH15] High-quality streamable free-viewpoint video</b> [project page] [pdf]

<b>[TVCG10] A point-cloud-based multiview stereo algorithm for free-viewpoint video</b> [pdf]

<b>[SIGGRAPH08] Markerless garment capture</b> [project page][pdf]

<b>[CGA07] Surface capture for performance-based animation</b> [pdf]

<b>[TVC05] Scalable 3d video of dynamic scenes</b> [pdf]

4. Photometric Stereo

<b>[ICCV11] Shading-based dynamic shape refinement from multi-view video under general illumination</b> [pdf]

<b>[SIGGRAPH_Asia09] Dynamic shape capture using multi-view photometric stereo</b> [project page] [pdf]

5. Template based Approaches

<b>[TOG13] On-set performance capture of multiple actors with a stereo camera</b> [pdf]

<b>[ECCV12] Full body performance capture under uncontrolled and varying illumination: A shading-based approach</b> [pdf]

<b>[CVPR11] Markerless motion capture of interacting characters using multi-view image segmentation</b> [pdf]

<b>[SIGGRAPH_Asia10] Video-based reconstruction of animatable human characters</b> [project page][pdf]

<b>[SIGGRAPH08] Performance capture from sparse multi-view video</b> [project page][pdf]

<b>[CVPR09] Motion capture using joint skeleton tracking and surface estimation</b> [pdf]

<b>[CVPR08] Robust fusion of dynamic shape and normal capture for high-quality reconstruction of time-varying geometry</b> [pdf]

6. Parametric Models

<b>[ICCV19] Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop</b> [pdf]

<b>[CVPR19] Expressive Body Capture: 3D Hands, Face, and Body from a Single Image</b> [project page] [pdf]

<b>[CVPR18 Oral] Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies</b> [project page] [pdf]

<b>[CVPR18] End-to-end recovery of human shape and pose</b> [project page] [code and data] [pdf]

<b>[CVPR17] Unite the people: Closing the loop between 3d and 2d human representations</b> [project page(code and data)] [pdf]

<b>[BMVC17] Indirect deep structured learning for 3D human body shape and pose prediction</b> [pdf]

<b>[ECCV16] Keep it SMPL: Automatic estimation of 3d human pose and shape from a single image</b> [code] [pdf]

<b>[SIGGRAPH_Asia15] SMPL: A skinned multi-person linear model</b> [project page][pdf]

<b>[TOG14] Mosh: Motion and shape capture from sparse markers</b> [project page] [pdf]

<b>[CVPR10] Multilinear pose and body shape estimation of dressed subjects from image sets</b> [pdf]

<b>[ICCV09] Estimating Human Shape and Pose from a Single Image</b> [pdf]

<b>[CVPR07] Detailed human shape and pose from images</b> [pdf]

<b>[SIGGRAPH05] SCAPE: shape completion and animation of people</b> [project page] [pdf]

<b>[CVIU01] Tracking and modeling people in video sequences</b> [pdf]

Single-view Reconstruction

<b>[ICCV19] Tex2Shape: Detailed Full Human Body Geometry From a Single Image</b> [pdf]

<b>[ICCV19] PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization</b> [pdf]

<b>[ICCV19] Moulding Humans: Non-parametric 3D Human Shape Estimation from Single Images</b> [pdf]

<b>[ICCV19] 3DPeople: Modeling the Geometry of Dressed Humans</b> [pdf]

<b>[ICCV19] DeepHuman: 3D Human Reconstruction From a Single Image</b> [pdf]

<b>[ICCV19] A Neural Network for Detailed Human Depth Estimation From a Single Image</b> [pdf]

<b>[ICCV19] Delving Deep Into Hybrid Annotations for 3D Human Recovery in the Wild</b> [pdf]

<b>[CVPR19 Oral] SiCloPe: Silhouette-Based Clothed People </b>[pdf]

<b>[CVPR19 Oral] Convolutional Mesh Regression for Single-Image Human Shape Reconstruction</b> [pdf]

<b>[CVPR19 Oral] Detailed Human Shape Estimation from a Single Image by Hierarchical mesh deformation</b> [pdf]

<b>[CVPR19] Learning to Reconstruct People in Clothing from a Single RGB Camera </b> [project page] [pdf]

<b>[CVPR19] SimulCap: Single-View Human Performance Capture with Cloth Simulation</b> [pdf]

<b>[CVPR19] Photo Wake-Up: 3D Character Animation from a Single Photo</b> [project page] [pdf]

<b>[TOG19] LiveCap: Real-time Human Performance Capture from Monocular Video</b> [project page] [pdf]

<b>[CVPR18] End-to-end recovery of human shape and pose</b> [project page] [code and data] [pdf]

<b>[3DV18] Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose</b> [pdf]

<b>[ICCV09] Estimating Human Shape and Pose from a Single Image</b> [pdf]

4D-Scans

<b>[SIGGRAPH17] ClothCap: seamless 4D clothing capture and retargeting</b> [project page] [pdf]

Datasets and Code

<b>MPI datasets and code</b>[Link]

<b>BUFF</b>[Link]

<b>Dynamic FAUST</b>[Link]

<b>CVSSP3D</b>[Link]

<b>Mixamo</b>[Link]

<b>Human3.6M</b>[Link]

<b>3DPeople</b>[Link]

<b>THuman</b>[Link]

<b>RenderPeople</b>[Link]

<b>CLOTH3D</b>[Link](not released yet)

<b>Multi-Garment Net</b>[Link]