Task-Specific Data Augmentation and Inference Processing for VIPriors Instance Segmentation Challenge

TL;DR

This paper introduces task-specific data augmentation and inference strategies to improve instance segmentation performance in data-scarce environments, demonstrated on the VIPriors challenge with competitive results.

Contribution

The paper proposes novel data augmentation and inference processing methods tailored for data-deficient instance segmentation tasks, leveraging visual inductive priors.

Findings

Achieved 0.531 AP@0.50:0.95 on VIPriors test set

Demonstrated effectiveness of task-specific strategies in data-limited scenarios

Validated approach with competitive results in the challenge

Abstract

Instance segmentation is applied widely in image editing, image analysis and autonomous driving, etc. However, insufficient data is a common problem in practical applications. The Visual Inductive Priors(VIPriors) Instance Segmentation Challenge has focused on this problem. VIPriors for Data-Efficient Computer Vision Challenges ask competitors to train models from scratch in a data-deficient setting, but there are some visual inductive priors that can be used. In order to address the VIPriors instance segmentation problem, we designed a Task-Specific Data Augmentation(TS-DA) strategy and Inference Processing(TS-IP) strategy. The main purpose of task-specific data augmentation strategy is to tackle the data-deficient problem. And in order to make the most of visual inductive priors, we designed a task-specific inference processing strategy. We demonstrate the applicability of proposed method on VIPriors Instance Segmentation Challenge. The segmentation model applied is Hybrid Task Cascade based detector on the Swin-Base based CBNetV2 backbone. Experimental results demonstrate that proposed method can achieve a competitive result on the test set of 2022 VIPriors Instance Segmentation Challenge, with 0.531 AP@0.50:0.95.