Ev-TTA: Test-Time Adaptation for Event-Based Object Recognition
This addresses the problem of domain shift for researchers and practitioners using event cameras in real-world applications, though it is an incremental improvement on existing adaptation methods.
The paper tackles performance deterioration in event-based object recognition under extreme conditions like fast motion or illumination changes by introducing Ev-TTA, a test-time adaptation algorithm that fine-tunes pre-trained classifiers online, achieving significant performance gains across various tasks without extensive retraining.
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.