CMC v2: Towards More Accurate COVID-19 Detection with Discriminative Video Priors

TL;DR

This paper introduces CMC v2, an improved COVID-19 detection method using video priors and a video transformer backbone, achieving top performance in a competitive challenge.

Contribution

The paper presents CMC v2, a novel approach that incorporates natural video priors and advanced training strategies for more accurate COVID-19 detection.

Findings

Ranked 1st in the COVID-19 competition

Achieved an average Macro F1 Score of 89.11%

Enhanced robustness and generalization of the model

Abstract

This paper presents our solution for the 2nd COVID-19 Competition, occurring in the framework of the AIMIA Workshop at the European Conference on Computer Vision (ECCV 2022). In our approach, we employ the winning solution last year which uses a strong 3D Contrastive Mixup Classifcation network (CMC v1) as the baseline method, composed of contrastive representation learning and mixup classification. In this paper, we propose CMC v2 by introducing natural video priors to COVID-19 diagnosis. Specifcally, we adapt a pre-trained (on video dataset) video transformer backbone to COVID-19 detection. Moreover, advanced training strategies, including hybrid mixup and cutmix, slicelevel augmentation, and small resolution training are also utilized to boost the robustness and the generalization ability of the model. Among 14 participating teams, CMC v2 ranked 1st in the 2nd COVID-19 Competition with an average Macro F1 Score of 89.11%.