Spiking Tucker Fusion Transformer for Audio-Visual Zero-Shot Learning

TL;DR

This paper introduces a novel Spiking Tucker Fusion Transformer that combines SNNs and transformers to improve audio-visual zero-shot learning by effectively integrating temporal and semantic information, achieving state-of-the-art results.

Contribution

The paper proposes a new Spiking Tucker Fusion Transformer with dynamic inference, global-local pooling, and a Tucker fusion module for multi-scale SNN and transformer integration in ZSL.

Findings

Achieves state-of-the-art performance on three benchmark datasets.

Significant harmonic mean improvements of 15.4%, 3.9%, and 14.9%.

Demonstrates effective fusion of temporal and semantic features.

Abstract

The spiking neural networks (SNNs) that efficiently encode temporal sequences have shown great potential in extracting audio-visual joint feature representations. However, coupling SNNs (binary spike sequences) with transformers (float-point sequences) to jointly explore the temporal-semantic information still facing challenges. In this paper, we introduce a novel Spiking Tucker Fusion Transformer (STFT) for audio-visual zero-shot learning (ZSL). The STFT leverage the temporal and semantic information from different time steps to generate robust representations. The time-step factor (TSF) is introduced to dynamically synthesis the subsequent inference information. To guide the formation of input membrane potentials and reduce the spike noise, we propose a global-local pooling (GLP) which combines the max and average pooling operations. Furthermore, the thresholds of the spiking neurons are dynamically adjusted based on semantic and temporal cues. Integrating the temporal and semantic information extracted by SNNs and Transformers are difficult due to the increased number of parameters in a straightforward bilinear model. To address this, we introduce a temporal-semantic Tucker fusion module, which achieves multi-scale fusion of SNN and Transformer outputs while maintaining full second-order interactions. Our experimental results demonstrate the effectiveness of the proposed approach in achieving state-of-the-art performance in three benchmark datasets. The harmonic mean (HM) improvement of VGGSound, UCF101 and ActivityNet are around 15.4\%, 3.9\%, and 14.9\%, respectively.