Skip to content

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization

Notifications You must be signed in to change notification settings

VisionXLab/Point2RBox-v3

Repository files navigation

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization

Abstract

Driven by the growing need for Oriented Object Detection (OOD), learning from point annotations under a weakly-supervised framework has emerged as a promising alternative to costly and laborious manual labeling. In this paper, we discuss two deficiencies in existing point-supervised methods: inefficient utilization and poor quality of pseudo labels. Therefore, we present Point2RBox-v3. At the core are two principles: \textbf{1) Progressive Label Assignment (PLA)}. It dynamically estimates instance sizes in a coarse yet intelligent manner at different stages of the training process, enabling the use of label assignment methods. \textbf{2) Prior-Guided Dynamic Mask Loss (PGDM-Loss)}. It is an enhancement of the Voronoi Watershed Loss from Point2RBox-v2, which overcomes the shortcomings of Watershed in its poor performance in sparse scenes and SAM's poor performance in dense scenes. To our knowledge, Point2RBox-v3 is the first model to employ dynamic pseudo labels for label assignment, and it creatively complements the advantages of SAM model with the watershed algorithm, which achieves excellent performance in both sparse and dense scenes. Our solution gives competitive performance, especially in scenarios with large variations in object size or sparse object occurrences: 66.09\%/56.86\%/41.28\%/46.40\%/19.60\%/45.96\% on DOTA-v1.0/DOTA-v1.5/DOTA-v2.0/DIOR/STAR/RSAR.

TODOs

  • Release the paper on arXiv.
  • Release the complete code.
  • Release the checkpoints.

Overview

  • Visual Comparison & Radar Evaluation.

Fig1

  • An Overview of Point2RBox-v3 and Pipeline.

arch

  • The process of Progressive Label Assignment (PLA).

arch

  • Comparison between watershed and SAM masks on DOTA-v1.0.

arch

Main Results

  • Detection performance of all categories and the mean AP50 on the DOTA-v1.0

arch

  • AP$_{50}$ comparisons on the DOTA-v1.0/1.5/2.0, DIOR, STAR, and RSAR datasets.
arch
  • AP$_{50}$ comparison on DOTA-v1.0/v1.5 under the partial weakly-supervised setting.

arch

Contact

If you have any questions about this paper or code, feel free to email me at zhangteng@sjtu.edu.cn. This ensures I can promptly notice and respond!

About

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •