Instance segmentation aims to predict the semantic and instance labels of each image pixel. It attracts more and more research interest as it provides more fine-grained information than object detection or semantic segmentation.
A recent study on arXiv.org proposes a brand new dataset that focuses on irregular shape instance segmentation, unlike most current methods, which focus on regularly shaped objects.
It consists of six sub-datasets, each of which represents scenes of a typical irregular shape, for example, strip shape, hollow shape, and mesh shape. The researchers also benchmark popular instance segmentation methods to reveal the drawbacks of popular methods on irregularly shaped objects.
A novel affinity-based instance segmentation baseline is proposed. It explicitly combines perception and reasoning to solve Arbitrary Shape Instance Segmentation and outperforms popular methods by a large margin.
In this paper, we introduce a brand new dataset to promote the study of instance segmentation for objects with irregular shapes. Our key observation is that though irregularly shaped objects widely exist in daily life and industrial scenarios, they received little attention in the instance segmentation field due to the lack of corresponding datasets. To fill this gap, we propose iShape, an irregular shape dataset for instance segmentation. iShape contains six sub-datasets with one real and five synthetics, each represents a scene of a typical irregular shape. Unlike most existing instance segmentation datasets of regular objects, iShape has many characteristics that challenge existing instance segmentation algorithms, such as large overlaps between bounding boxes of instances, extreme aspect ratios, and large numbers of connected components per instance. We benchmark popular instance segmentation methods on iShape and find their performance drop dramatically. Hence, we propose an affinity-based instance segmentation algorithm, called ASIS, as a stronger baseline. ASIS explicitly combines perception and reasoning to solve Arbitrary Shape Instance Segmentation including irregular objects. Experimental results show that ASIS outperforms the state-of-the-art on iShape. Dataset and code are available at this https URL
Research paper: Yang, L., “iShape: A First Step Towards Irregular Shape Instance Segmentation”, 2021. Link: https://arxiv.org/abs/2109.15068