Interactive Instance Annotation with Siamese Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Xiang, Li, Ruotong, Yi, Mengjun, XU, Baile, Shen, Furao, Zhao, Jian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912362067394560
author Xu, Xiang
Li, Ruotong
Yi, Mengjun
XU, Baile
Shen, Furao
Zhao, Jian
author_facet Xu, Xiang
Li, Ruotong
Yi, Mengjun
XU, Baile
Shen, Furao
Zhao, Jian
contents Annotating instance masks is time-consuming and labor-intensive. A promising solution is to predict contours using a deep learning model and then allow users to refine them. However, most existing methods focus on in-domain scenarios, limiting their effectiveness for cross-domain annotation tasks. In this paper, we propose SiamAnno, a framework inspired by the use of Siamese networks in object tracking. SiamAnno leverages one-shot learning to annotate previously unseen objects by taking a bounding box as input and predicting object boundaries, which can then be adjusted by annotators. Trained on one dataset and tested on another without fine-tuning, SiamAnno achieves state-of-the-art (SOTA) performance across multiple datasets, demonstrating its ability to handle domain and environment shifts in cross-domain tasks. We also provide more comprehensive results compared to previous work, establishing a strong baseline for future research. To our knowledge, SiamAnno is the first model to explore Siamese architecture for instance annotation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_03184
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Interactive Instance Annotation with Siamese Networks
Xu, Xiang
Li, Ruotong
Yi, Mengjun
XU, Baile
Shen, Furao
Zhao, Jian
Computer Vision and Pattern Recognition
Annotating instance masks is time-consuming and labor-intensive. A promising solution is to predict contours using a deep learning model and then allow users to refine them. However, most existing methods focus on in-domain scenarios, limiting their effectiveness for cross-domain annotation tasks. In this paper, we propose SiamAnno, a framework inspired by the use of Siamese networks in object tracking. SiamAnno leverages one-shot learning to annotate previously unseen objects by taking a bounding box as input and predicting object boundaries, which can then be adjusted by annotators. Trained on one dataset and tested on another without fine-tuning, SiamAnno achieves state-of-the-art (SOTA) performance across multiple datasets, demonstrating its ability to handle domain and environment shifts in cross-domain tasks. We also provide more comprehensive results compared to previous work, establishing a strong baseline for future research. To our knowledge, SiamAnno is the first model to explore Siamese architecture for instance annotation.
title Interactive Instance Annotation with Siamese Networks
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.03184