Exploring 2D Data Augmentation for 3D Monocular Object Detection

04/21/2021
by   Sugirtha T, et al.
7

Data augmentation is a key component of CNN based image recognition tasks like object detection. However, it is relatively less explored for 3D object detection. Many standard 2D object detection data augmentation techniques do not extend to 3D box. Extension of these data augmentations for 3D object detection requires adaptation of the 3D geometry of the input scene and synthesis of new viewpoints. This requires accurate depth information of the scene which may not be always available. In this paper, we evaluate existing 2D data augmentations and propose two novel augmentations for monocular 3D detection without a requirement for novel view synthesis. We evaluate these augmentations on the RTM3D detection model firstly due to the shorter training times . We obtain a consistent improvement by 4 cars,  1.8 baseline on KITTI car detection dataset. We also demonstrate a rigorous evaluation of the mAP scores by re-weighting them to take into account the class imbalance in the KITTI validation dataset.

READ FULL TEXT

page 1

page 4

page 5

page 6

research
06/25/2022

Self-Supervised 3D Monocular Object Detection by Recycling Bounding Boxes

Modern object detection architectures are moving towards employing self-...
research
04/03/2020

Quantifying Data Augmentation for LiDAR based 3D Object Detection

In this work, we shed light on different data augmentation techniques co...
research
04/02/2020

Improving 3D Object Detection through Progressive Population Based Augmentation

Data augmentation has been widely adopted for object detection in 3D poi...
research
07/27/2020

Part-Aware Data Augmentation for 3D Object Detection in Point Cloud

Data augmentation has greatly contributed to improving the performance i...
research
05/22/2023

Real-Aug: Realistic Scene Synthesis for LiDAR Augmentation in 3D Object Detection

Data and model are the undoubtable two supporting pillars for LiDAR obje...
research
10/09/2022

Data augmentation for NeRF: a geometric consistent solution based on view morphing

NeRF aims to learn a continuous neural scene representation by using a f...
research
12/06/2018

ROI-10D: Monocular Lifting of 2D Detection to 6D Pose and Metric Shape

We present a deep learning method for end-to-end monocular 3D object det...

Please sign up or login with your details

Forgot password? Click here to reset