Documents
Poster
Poster
AEAM3D: ADVERSE ENVIRONMENT-ADAPTIVE MONOCULAR 3D OBJECT DETECTION VIA FEATURE EXTRACTION REGULARIZATION
- Citation Author(s):
- Submitted by:
- Yixin Lei
- Last updated:
- 30 March 2024 - 2:52am
- Document Type:
- Poster
- Document Year:
- 2024
- Event:
- Presenters:
- Jinyuan Liu
- Paper Code:
- IVMSP-P14.5
- Categories:
- Keywords:
- Log in to post comments
3D object detection plays a crucial role in intelligent vision systems. Detection in the open world inevitably encounters various adverse scenes while most of existing methods fail in these scenes. To address this issue, this paper proposes a monocular 3D detection model, termed AEAM3D, which effectively mitigates the degradation of detection performance in various harsh environments. Additionally, we assemble a new adverse 3D object detection dataset encompassing some challenging scenes, including rainy, foggy, and low light
weather conditions. Experimental results demonstrate that our proposed method outperforms current state-of-the-art approaches by an average of 3.12% in terms of AP_R40 for car category across adverse environments.