One-Shot Instance Segmentation

By Claudio Michaelis, Ivan Ustyuzhaninov, Matthias Bethge, Alexander S. Ecker. 2018

99 citations recorded by the source corpus.

We tackle the problem of one-shot instance segmentation: Given an example image of a novel, previously unknown object category, find and segment all objects of this category within a complex scene. To address this challenging new task, we propose Siamese Mask R-CNN. It extends Mask R-CNN by a Siamese backbone encoding both reference image and scene, allowing it to target detection and segmentation towards the reference category. We demonstrate empirical results on MS Coco highlighting challenges of the one-shot setting: while transferring knowledge about instance segmentation to novel object categories works very well, targeting the detection network towards the reference category appears to be more difficult. Our work provides a first strong baseline for one-shot instance segmentation and will hopefully inspire further research into more powerful and flexible scene analysis algorithms. Code is available at: this https URL

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. B. Boom
2. Ronald Poppe
3. Dan Andrei Ganea
4. Changxin Gao
5. Guisong Xia

In Computer Vision and Pattern Recognition

1. Guisong Xia
2. B. Boom
3. Ronald Poppe
4. Dan Andrei Ganea
5. Yuanqing Li
6. Jiarong Ou
7. Changxin Gao
8. Zhihao Liang

Citations in the corpus

• Incremental Few-Shot Instance Segmentation (2021, methodology): “Few works have addressed FSIS [8, 21, 39].”
• A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, unknown): “8 Siamese Mask R-CNN [268] Arxiv’18 ResNet-50 Metric ✓ ✓ ✓ 14.”
• FGN: Fully Guided Network for Few-Shot Instance Segmentation (2020, methodology): “Most closely related to ours, the works in [21, 30] consider FSIS by constructing guided networks upon Mask R-CNN.”
• Image Segmentation Using Text and Image Prompts (2021, unknown): “In one-shot instance segmentation [42], instead of a binary match/non-match prediction, individual object instances are segmented.”