Contrastive lift: 3D object instance segmentation by slow-fast contrastive fusion
Instance segmentation in 3D is a challenging task due to the lack of large-scale annotated datasets. In this paper, we show that this task can be addressed effectively by leveraging instead 2D pre-trained models for instance segmentation. We propose a novel approach to lift 2D segments to 3D and fus...
Главные авторы: | Bhalgat, Y, Laina, I, Henriques, JF, Zisserman, A, Vedaldi, A |
---|---|
Формат: | Conference item |
Язык: | English |
Опубликовано: |
Curran Associates
2024
|
Схожие документы
-
3D-aware instance segmentation and tracking in egocentric videos
по: Bhalgat, Y, и др.
Опубликовано: (2025) -
Instance Segmentation of Shrimp Based on Contrastive Learning
по: Heng Zhou, и др.
Опубликовано: (2023-06-01) -
N2F2: hierarchical scene understanding with nested neural feature fields
по: Bhalgat, YS, и др.
Опубликовано: (2024) -
Instance-Level Contrastive Learning for Weakly Supervised Object Detection
по: Ming Zhang, и др.
Опубликовано: (2022-10-01) -
Unsupervised part discovery from contrastive reconstruction
по: Choudhury, S, и др.
Опубликовано: (2021)