computer-vision-3

3D Shape Reconstruction from Videos

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in 3d-shape-reconstruction-from-videos-3

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find 3d-shape-reconstruction-from-videos-3 models and implementations

Datasets

No datasets available.

Subtasks

DeepFake Detection

Most implemented papers

Reconstructing Animatable Categories from Videos

Deva Ramanan, Chaoyang Wang, Dinesh Reddy Narapureddy, Gengshan Yang•Tue May 09 2023

RAC, a method to build category-level 3D models from monocular videos, disentangling variations over instances and motion over time is presented, using 3D background models to disentangle objects from the background.

46

Content

0

Paper Graph

LASR: Learning Articulated Shape Reconstruction from a Monocular Video

Ce Liu, Huiwen Chang, W. Freeman, Deva Ramanan, Forrester Cole, Daniel Vlasic, V. Jampani, Deqing Sun, Gengshan Yang•Wed May 05 2021

This work introduces a template-free approach to learn 3D shapes from a single video with an analysis-by-synthesis strategy that forward-renders object silhouette, optical flow, and pixel values to compare with video observations, which generates gradients to adjust the camera, shape and motion parameters.

137 0

Paper Graph

ViSER: Video-Specific Surface Embeddings for Articulated 3D Shape Reconstruction

Ce Liu, Deva Ramanan, Forrester Cole, Daniel Vlasic, Varun Jampani, Deqing Sun, Gengshan Yang•Thu Dec 31 2020

Experimental results show that ViSER compares favorably against prior work on challenging videos of humans with loose clothing and unusual poses as well as animals videos from DAVIS and YTVOS.

97 0

Paper Graph

BANMo: Building Animatable 3D Neural Models from Many Casual Videos

A. Vedaldi, Deva Ramanan, N. Neverova, H. Joo, Minh Vo, Gengshan Yang•Wed Dec 22 2021

This work aims to create high-fidelity, articulated 3D models from many casual RGB videos in a differentiable rendering framework, and introduces neural blend skinning models that allow for differentiable and invertible articulated deformations.

227 0

Paper Graph

DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation

Bo Dai, Lei Yang, Ailing Zeng, Xu Ju, Ruiyuan Gao, Xizhou Zhu, Qianglin Xu•Tue Mar 15 2022

DeciWatch introduces a simple yet effective sample-denoise-recover framework that only watches sparsely sampled frames, taking advantage of the continuity of human motions and the lightweight pose representation.

55 0

Paper Graph

Adding a benchmark result helps the community track progress.