WSESeg: Introducing a Dataset for the Segmentation of Winter Sports Equipment with a Baseline for Interactive Segmentation

TL;DR

This paper introduces WSESeg, a new dataset for winter sports equipment segmentation, and evaluates interactive segmentation models and adaptation methods to improve labeling efficiency and accuracy.

Contribution

It provides a novel dataset for winter sports equipment segmentation and explores online adaptation techniques to enhance interactive segmentation performance.

Findings

Adaptation methods significantly reduce failure rate and number of clicks.

Models generalize well to new winter sports equipment categories.

Interactive segmentation becomes more efficient with online adaptation.

Abstract

In this paper we introduce a new dataset containing instance segmentation masks for ten different categories of winter sports equipment, called WSESeg (Winter Sports Equipment Segmentation). Furthermore, we carry out interactive segmentation experiments on said dataset to explore possibilities for efficient further labeling. The SAM and HQ-SAM models are conceptualized as foundation models for performing user guided segmentation. In order to measure their claimed generalization capability we evaluate them on WSESeg. Since interactive segmentation offers the benefit of creating easily exploitable ground truth data during test-time, we are going to test various online adaptation methods for the purpose of exploring potentials for improvements without having to fine-tune the models explicitly. Our experiments show that our adaptation methods drastically reduce the Failure Rate (FR) and Number of Clicks (NoC) metrics, which generally leads faster to better interactive segmentation results.