Saved in:
Bibliographic Details
Main Authors: Jhang, Jin-Cheng, Tu, Tao, Wang, Fu-En, Zhang, Ke, Sun, Min, Kuo, Cheng-Hao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.11412
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929632653082624
author Jhang, Jin-Cheng
Tu, Tao
Wang, Fu-En
Zhang, Ke
Sun, Min
Kuo, Cheng-Hao
author_facet Jhang, Jin-Cheng
Tu, Tao
Wang, Fu-En
Zhang, Ke
Sun, Min
Kuo, Cheng-Hao
contents The field of indoor monocular 3D object detection is gaining significant attention, fueled by the increasing demand in VR/AR and robotic applications. However, its advancement is impeded by the limited availability and diversity of 3D training data, owing to the labor-intensive nature of 3D data collection and annotation processes. In this paper, we present V-MIND (Versatile Monocular INdoor Detector), which enhances the performance of indoor 3D detectors across a diverse set of object classes by harnessing publicly available large-scale 2D datasets. By leveraging well-established monocular depth estimation techniques and camera intrinsic predictors, we can generate 3D training data by converting large-scale 2D images into 3D point clouds and subsequently deriving pseudo 3D bounding boxes. To mitigate distance errors inherent in the converted point clouds, we introduce a novel 3D self-calibration loss for refining the pseudo 3D bounding boxes during training. Additionally, we propose a novel ambiguity loss to address the ambiguity that arises when introducing new classes from 2D datasets. Finally, through joint training with existing 3D datasets and pseudo 3D bounding boxes derived from 2D datasets, V-MIND achieves state-of-the-art object detection performance across a wide range of classes on the Omni3D indoor dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2412_11412
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations
Jhang, Jin-Cheng
Tu, Tao
Wang, Fu-En
Zhang, Ke
Sun, Min
Kuo, Cheng-Hao
Computer Vision and Pattern Recognition
The field of indoor monocular 3D object detection is gaining significant attention, fueled by the increasing demand in VR/AR and robotic applications. However, its advancement is impeded by the limited availability and diversity of 3D training data, owing to the labor-intensive nature of 3D data collection and annotation processes. In this paper, we present V-MIND (Versatile Monocular INdoor Detector), which enhances the performance of indoor 3D detectors across a diverse set of object classes by harnessing publicly available large-scale 2D datasets. By leveraging well-established monocular depth estimation techniques and camera intrinsic predictors, we can generate 3D training data by converting large-scale 2D images into 3D point clouds and subsequently deriving pseudo 3D bounding boxes. To mitigate distance errors inherent in the converted point clouds, we introduce a novel 3D self-calibration loss for refining the pseudo 3D bounding boxes during training. Additionally, we propose a novel ambiguity loss to address the ambiguity that arises when introducing new classes from 2D datasets. Finally, through joint training with existing 3D datasets and pseudo 3D bounding boxes derived from 2D datasets, V-MIND achieves state-of-the-art object detection performance across a wide range of classes on the Omni3D indoor dataset.
title V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.11412