Oct 16, 2020 · SF-UDA^{3D} outperforms both previous domain adaptation techniques based on features alignment and state-of-the-art 3D object detection methods ...
GPL-3.0 license. Official implementation of the 3DV 2020 paper "SF-UDA-3D: Source-Free Unsupervised Domain Adaptation for LiDAR-based 3D Object Detection".
This paper proposes SF-UDA3D, the first Source-Free Unsupervised Domain Adaptation (SF-UDA) framework to domain-adapt the state-of-the-art PointRCNN 3D ...
Results show that xMUDA significantly improves over uni-modal UDA in all tested scenarios and complements state-of-the-art UDA techniques. Saltori et al.
Missing: UDA3D: | Show results with:UDA3D:
Oct 16, 2020 · SF-UDA3D outperforms both previous domain adaptation techniques based on features alignment and state-of-the-art 3D object detection methods ...
Sf-uda 3d: Source-free unsupervised domain adaptation for lidar-based 3d object detection. Cristiano Saltori, Stëphane Lathuiliëre, Nicu Sebe, Elisa Ricci ...
Missing: UDA3D: | Show results with:UDA3D:
SF-UDA3D: Source-Free Unsupervised Domain Adaptation for LiDAR-Based 3D Object Detection. Pseudo-annotations, reversible scale-transformations and motion ...
We propose a single-frame approach for source-free, un- supervised domain adaptation of LiDAR-based 3D object detectors that uses class prototypes to mitigate ...
Sf-uda3d: Source-free unsupervised domain adaptation for lidar-based 3d object detection. C Saltori; S Lathuili'ere; N Sebe; E Ricci; F Galasso. Pseudo-labeling ...
In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly ...
Missing: UDA3D: | Show results with:UDA3D: