Temporal Bilinear Encoding Network of Audio-Visual Features at Low Sampling Rates
This work addresses the problem of efficient video classification for researchers and practitioners by enabling high performance with low sampling rates and reduced computational demands.
This paper introduces Temporal Bilinear Encoding Networks (TBEN) to classify videos using audio-visual information sampled at 1 frame per second. TBEN achieves a new state-of-the-art hit@1 of 47.95% on the FGA240 dataset and a hit@1 of 91.03% on UCF101, demonstrating competitive accuracy with significantly reduced computational resources.
Current deep learning based video classification architectures are typically trained end-to-end on large volumes of data and require extensive computational resources. This paper aims to exploit audio-visual information in video classification with a 1 frame per second sampling rate. We propose Temporal Bilinear Encoding Networks (TBEN) for encoding both audio and visual long range temporal information using bilinear pooling and demonstrate bilinear pooling is better than average pooling on the temporal dimension for videos with low sampling rate. We also embed the label hierarchy in TBEN to further improve the robustness of the classifier. Experiments on the FGA240 fine-grained classification dataset using TBEN achieve a new state-of-the-art (hit@1=47.95%). We also exploit the possibility of incorporating TBEN with multiple decoupled modalities like visual semantic and motion features: experiments on UCF101 sampled at 1 FPS achieve close to state-of-the-art accuracy (hit@1=91.03%) while requiring significantly less computational resources than competing approaches for both training and prediction.