CaltechAUTHORS
  A Caltech Library Service

FreeSOLO: Learning to Segment Objects without Annotations

Wang, Xinlong and Yu, Zhiding and De Mello, Shalini and Kautz, Jan and Anandkumar, Anima and Shen, Chunhua and Alvarez, Jose M. (2022) FreeSOLO: Learning to Segment Objects without Annotations. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20220714-224614512

[img] PDF - Submitted Version
See Usage Policy.

6MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20220714-224614512

Abstract

Instance segmentation is a fundamental vision task that aims to recognize and segment each object in an image. However, it requires costly annotations such as bounding boxes and segmentation masks for learning. In this work, we propose a fully unsupervised learning method that learns class-agnostic instance segmentation without any annotations. We present FreeSOLO, a self-supervised instance segmentation framework built on top of the simple instance segmentation method SOLO. Our method also presents a novel localization-aware pre-training framework, where objects can be discovered from complicated scenes in an unsupervised manner. FreeSOLO achieves 9.8% AP_{50} on the challenging COCO dataset, which even outperforms several segmentation proposal methods that use manual annotations. For the first time, we demonstrate unsupervised class-agnostic instance segmentation successfully. FreeSOLO's box localization significantly outperforms state-of-the-art unsupervised object detection/discovery methods, with about 100% relative improvements in COCO AP. FreeSOLO further demonstrates superiority as a strong pre-training method, outperforming state-of-the-art self-supervised pre-training methods by +9.8% AP when fine-tuning instance segmentation with only 5% COCO masks. Code is available at: github.com/NVlabs/FreeSOLO


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
https://doi.org/10.48550/arXiv.2202.12181arXivDiscussion Paper
ORCID:
AuthorORCID
Anandkumar, Anima0000-0002-6974-6797
Additional Information:Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) Part of this work was done when XW was an intern at NVIDIA, and CS was with The Univerity of Adelaide.
Record Number:CaltechAUTHORS:20220714-224614512
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20220714-224614512
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:115596
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:15 Jul 2022 23:27
Last Modified:15 Jul 2022 23:27

Repository Staff Only: item control page