Our paper has been accepted in “Asian Conference on Computer Vision (ACCV) 2024” Main Conference
Our paper on the multi-task approach to self-supervised depth estimation and semantic segmentation has been accepted in ACCV 2024, the 17th Asian Conference on Computer Vision, a premier forum for researchers and developers to present and discuss the latest advancements in computer vision technologies. Our research focuses on performing both depth estimation and semantic segmentation simultaneously and consistently in both daytime and nighttime scenarios, all without the need for labeled data. In the future, we aim to enhance our model’s generalization ability to ensure robustness across various weather conditions beyond the standard day-night scenarios.
17th Asian Conference on Computer Vision
DepthSegNet24: A Label-Free Model for Robust Day-Night Depth and Semantics
Phan Thi Huyen Thanh 1, The Hiep Nguyen 2, Minh Huy Vu Nguyen 2, Trung Thai Tran 2, Tran Vu Pham 2, Duc Dung Nguyen 2, Truong Vinh Truong Duy 1, Natori Naotake 1
1: Aisin Corporation
2: Hochiminh City University of Technology (HCMUT)