Paper Image

Image-based 3D detection using geometry

Published on:

17 August 2023

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Tao Tu,

Shun-Po Chuang,

Yu-Lun Liu,

Cheng Sun,

Ke Zhang,

Donna Roy,

Cheng-Hao Kuo,

Min Sun

Bullets

Key Details

Uses images to induce geometry in a 3D voxel space

Geometry prediction reduces confusion from empty voxels

Achieves state-of-the-art image-based detection accuracy

Shows value of inducing geometry from images

Enables image-based methods to beat point clouds

AI generated summary

Image-based 3D detection using geometry

This paper proposes a 3D object detection method called ImGeoNet that uses images to model a 3D space. It learns to predict geometry from the images, reducing confusion from empty space voxels. This image-based approach achieved state-of-the-art accuracy, showing the value of inducing geometry from images.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up