Saved in:
Bibliographic Details
Main Authors: Li, Shichao, Huang, Xijie, Liu, Zechun, Cheng, Kwang-Ting
Format: Preprint
Published: 2021
Subjects:
Online Access:https://arxiv.org/abs/2111.12924
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913391399927808
author Li, Shichao
Huang, Xijie
Liu, Zechun
Cheng, Kwang-Ting
author_facet Li, Shichao
Huang, Xijie
Liu, Zechun
Cheng, Kwang-Ting
contents We present a new learning-based framework S-3D-RCNN that can recover accurate object orientation in SO(3) and simultaneously predict implicit rigid shapes from stereo RGB images. For orientation estimation, in contrast to previous studies that map local appearance to observation angles, we propose a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs). This approach features a deep model that transforms perceived intensities from one or two views to object part coordinates to achieve direct egocentric object orientation estimation in the camera coordinate system. To further achieve finer description inside 3D bounding boxes, we investigate the implicit shape estimation problem from stereo images. We model visible object surfaces by designing a point-based representation, augmenting IGRs to explicitly address the unseen surface hallucination problem. Extensive experiments validate the effectiveness of the proposed IGRs, and S-3D-RCNN achieves superior 3D scene understanding performance. We also designed new metrics on the KITTI benchmark for our evaluation of implicit shape estimation.
format Preprint
id arxiv_https___arxiv_org_abs_2111_12924
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Joint stereo 3D object detection and implicit surface reconstruction
Li, Shichao
Huang, Xijie
Liu, Zechun
Cheng, Kwang-Ting
Computer Vision and Pattern Recognition
Graphics
Robotics
We present a new learning-based framework S-3D-RCNN that can recover accurate object orientation in SO(3) and simultaneously predict implicit rigid shapes from stereo RGB images. For orientation estimation, in contrast to previous studies that map local appearance to observation angles, we propose a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs). This approach features a deep model that transforms perceived intensities from one or two views to object part coordinates to achieve direct egocentric object orientation estimation in the camera coordinate system. To further achieve finer description inside 3D bounding boxes, we investigate the implicit shape estimation problem from stereo images. We model visible object surfaces by designing a point-based representation, augmenting IGRs to explicitly address the unseen surface hallucination problem. Extensive experiments validate the effectiveness of the proposed IGRs, and S-3D-RCNN achieves superior 3D scene understanding performance. We also designed new metrics on the KITTI benchmark for our evaluation of implicit shape estimation.
title Joint stereo 3D object detection and implicit surface reconstruction
topic Computer Vision and Pattern Recognition
Graphics
Robotics
url https://arxiv.org/abs/2111.12924