TY - GEN
T1 - 3×2
T2 - 18th European Conference on Computer Vision, ECCV 2024
AU - Thai, Anh
AU - Wang, Weiyao
AU - Tang, Hao
AU - Stojanov, Stefan
AU - Rehg, James M.
AU - Feiszli, Matt
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - 3D object part segmentation is essential in computer vision applications. While substantial progress has been made in 2D object part segmentation, the 3D counterpart has received less attention, in part due to the scarcity of annotated 3D datasets, which are expensive to collect. In this work, we propose to leverage a few annotated 3D shapes or richly annotated 2D datasets to perform 3D object part segmentation. We present our novel approach, termed 3-By-2 that achieves SOTA performance on different benchmarks with various granularity levels. By using features from pretrained foundation models and exploiting semantic and geometric correspondences, we are able to overcome the challenges of limited 3D annotations. Our approach leverages available 2D labels, enabling effective 3D object part segmentation. Our method 3-By-2 can accommodate various part taxonomies and granularities, demonstrating part label transfer ability across different object categories. Project website: https://ngailapdi.github.io/projects/3by2/.
AB - 3D object part segmentation is essential in computer vision applications. While substantial progress has been made in 2D object part segmentation, the 3D counterpart has received less attention, in part due to the scarcity of annotated 3D datasets, which are expensive to collect. In this work, we propose to leverage a few annotated 3D shapes or richly annotated 2D datasets to perform 3D object part segmentation. We present our novel approach, termed 3-By-2 that achieves SOTA performance on different benchmarks with various granularity levels. By using features from pretrained foundation models and exploiting semantic and geometric correspondences, we are able to overcome the challenges of limited 3D annotations. Our approach leverages available 2D labels, enabling effective 3D object part segmentation. Our method 3-By-2 can accommodate various part taxonomies and granularities, demonstrating part label transfer ability across different object categories. Project website: https://ngailapdi.github.io/projects/3by2/.
UR - http://www.scopus.com/inward/record.url?scp=85206375680&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85206375680&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-72920-1_9
DO - 10.1007/978-3-031-72920-1_9
M3 - Conference contribution
AN - SCOPUS:85206375680
SN - 9783031729195
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 149
EP - 166
BT - Computer Vision – ECCV 2024 - 18th European Conference, Proceedings
A2 - Leonardis, Aleš
A2 - Ricci, Elisa
A2 - Roth, Stefan
A2 - Russakovsky, Olga
A2 - Sattler, Torsten
A2 - Varol, Gül
PB - Springer
Y2 - 29 September 2024 through 4 October 2024
ER -