Year: 2025
Type: conference-paper
Abstract: Instance segmentation models are crucial for precise object detection but often require expensive pixel-wise mask annotations.This paper studies the impact of combining bounding box and mask annotations in semi-supervised segmentation.We propose a method that leverages from both types of labeled data within a unified training framework.Through experiments on YOLO (convolution-based) and DETR (transformer-based) architectures, we demonstrate that balancing these annotation types significantly enhances performance while reducing labeling costs, particularly in terms of manual annotation time.Additionally, we evaluate few-shot and zero-shot scenarios, further highlighting the flexibility and efficiency of our method for budget-constrained segmentation tasks.
Source: Computer Science Research Notes
DOI: 10.24132/csrn.2025-9
Authors: ENGIE Lab CRIGEN, Daniil Tolstykh, Dmitriy Slutskiy, ENGIE Lab CRIGEN
Institution: Engie (France)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4412763395