Ev-TTA: Test-Time Adaptation for Event-Based Object Recognition

TL;DR

Ev-TTA is a test-time adaptation method for event-based object recognition that improves performance under extreme conditions by online fine-tuning using spatio-temporal event characteristics.

Contribution

It introduces a novel test-time adaptation algorithm that leverages event data's temporal and spatial features to handle domain shifts without extensive retraining.

Findings

Significant performance improvements across various event-based recognition tasks.

Effective noise handling under extreme illumination conditions.

Applicable to both classification and regression tasks.

Abstract

We introduce Ev-TTA, a simple, effective test-time adaptation algorithm for event-based object recognition. While event cameras are proposed to provide measurements of scenes with fast motions or drastic illumination changes, many existing event-based recognition algorithms suffer from performance deterioration under extreme conditions due to significant domain shifts. Ev-TTA mitigates the severe domain gaps by fine-tuning the pre-trained classifiers during the test phase using loss functions inspired by the spatio-temporal characteristics of events. Since the event data is a temporal stream of measurements, our loss function enforces similar predictions for adjacent events to quickly adapt to the changed environment online. Also, we utilize the spatial correlations between two polarities of events to handle noise under extreme illumination, where different polarities of events exhibit distinctive noise distributions. Ev-TTA demonstrates a large amount of performance gain on a wide range of event-based object recognition tasks without extensive additional training. Our formulation can be successfully applied regardless of input representations and further extended into regression tasks. We expect Ev-TTA to provide the key technique to deploy event-based vision algorithms in challenging real-world applications where significant domain shift is inevitable.