Carview!

CARVIEW

MOTORHOMES

Select Language

HTTP/2 301 server: GitHub.com content-type: text/html location: https://yixchen.github.io/holisticpp/ x-github-request-id: 79EE:2C10E1:955D8E:A7892A:6952ECDC accept-ranges: bytes age: 0 date: Mon, 29 Dec 2025 21:04:29 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210043-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767042269.040475,VS0,VE197 vary: Accept-Encoding x-fastly-request-id: ce1fc9c74279e39912fe2fc0f81e6e2cd7f5f96c content-length: 162 HTTP/2 200 server: GitHub.com content-type: text/html; charset=utf-8 last-modified: Wed, 29 Nov 2023 01:36:14 GMT access-control-allow-origin: * strict-transport-security: max-age=31556952 etag: W/"6566958e-23a5" expires: Mon, 29 Dec 2025 21:14:29 GMT cache-control: max-age=600 content-encoding: gzip x-proxy-cache: MISS x-github-request-id: C76D:15317B:950D52:A7381B:6952ECDD accept-ranges: bytes age: 0 date: Mon, 29 Dec 2025 21:04:29 GMT via: 1.1 varnish x-served-by: cache-bom-vanm7210043-BOM x-cache: MISS x-cache-hits: 0 x-timer: S1767042269.251570,VS0,VE217 vary: Accept-Encoding x-fastly-request-id: 4e1417b037664a2889f8b820662543bad7cbba77 content-length: 3008 Holistic_SceneHuman

Abstract

We propose a new 3D holistic⁺⁺ scene understanding problem, which jointly tackles two tasks from a single-view image: (i) holistic scene parsing and reconstruction-3D estimations of object bounding boxes, camera pose, and room layout, and (ii) 3D human pose estimation. The intuition behind is to leverage the coupled nature of these two tasks to improve the granularity and performance of scene understanding. We propose to exploit two critical and essential connections between these two tasks: (i) human-object interaction(HOI) to model the fine-grained relations between agents and objects in the scene, and (ii) physical commonsense to model the physical plausibility of the reconstructed scene. The optimal configuration of the 3D scene, represented by a parse graph, is inferred using Markov Chain Monte Carlo(MCMC), which efficiently traverses through the non-differentiable joint solution space. Experimental results demonstrate that the proposed algorithm significantly improves the performance of the two tasks on three datasets, showing an improved generalization ability.

Video

Live Demo

Note that the 3D scene reconstruction in this live demo is pre-computed to reach real-time 3D pose estimation performance. The CAD models are placed based on estimated object category, location, and size information for better visualization.

Paper

Holistic⁺⁺ Scene Understanding: Single-view 3D Holistic Scene Parsing and Human Pose Estimation with Human-Object Interaction and Physical Commensense
Yixin Chen^*, Siyuan Huang^*, Tao Yuan, Siyuan Qi, Yixin Zhu, Song-Chun Zhu
The IEEE International Conference on Computer Vision (ICCV) 2019
(^* indicates equal contribution.)
Paper / Supplementary / Code / Bibtex

Team

Yixin Chen¹

Siyuan Huang¹

Tao Yuan¹

Siyuan Qi²

Yixin Zhu²

Song-Chun Zhu^1,2

¹ Department of Statistics, UCLA

² Department of Computer Science, UCLA

Bibtex

@inProceedings{chen2019holisticpp,
title={Holistic++ Scene Understanding: Single-view 3D Holistic Scene Parsing and Human Pose Estimation with Human-Object Interaction and Physical Commonsense},
author = {Chen, Yixin and Huang, Siyuan and Yuan, Tao and Qi, Siyuan and Zhu, Yixin and Zhu, Song-Chun},
booktitle={The IEEE International Conference on Computer Vision (ICCV),
year={2019}
}

Original Source | Taken Source