Cell Behavior Video Classification Challenge, a benchmark for computer vision methods in time-lapse microscopy

TL;DR

This paper introduces the Cell Behavior Video Classification Challenge to benchmark computer vision methods for classifying complex cellular behaviors in microscopy videos, highlighting different approaches and their effectiveness.

Contribution

It organizes a large-scale benchmark comparing tracking-based, deep learning, and hybrid methods for cellular video classification, advancing the evaluation of techniques in this domain.

Findings

Deep learning methods show promising results in end-to-end classification.

Ensembling tracking and image features improves accuracy.

The benchmark reveals strengths and limitations of current approaches.

Abstract

The classification of microscopy videos capturing complex cellular behaviors is crucial for understanding and quantifying the dynamics of biological processes over time. However, it remains a frontier in computer vision, requiring approaches that effectively model the shape and motion of objects without rigid boundaries, extract hierarchical spatiotemporal features from entire image sequences rather than static frames, and account for multiple objects within the field of view. To this end, we organized the Cell Behavior Video Classification Challenge (CBVCC), benchmarking 35 methods based on three approaches: classification of tracking-derived features, end-to-end deep learning architectures to directly learn spatiotemporal features from the entire video sequence without explicit cell tracking, or ensembling tracking-derived with image-derived features. We discuss the results achieved by the participants and compare the potential and limitations of each approach, serving as a basis to foster the development of computer vision methods for studying cellular dynamics.