arXiv subjects

Danilo Pena

Publications and source records attributed to Danilo Pena.

3 recordsLinked to original sources

arXiv2026

Efficient Audio-Visual Event Recognition via Knowledge Distillation and Dynamic INT8 Quantization of a Hybrid Cross-Attention Network

Audio-visual event recognition (AVER) has achieved significant performance improvements through transformer-based multimodal architectures. However, the high computational complexity, large memory footprint, and inference cost of these models hinder their deployment on edge and resource-constrained devices. This paper presents an efficient compression framework for hybrid cross-attention-based audiovisual event recognition by combining architectural model compression, knowledge distillation, and dynamic INT8 quantization. A high-capacity teacher model integrates VideoMAE for visual representation learning, the Audio Spectrogram Transformer (AST) for audio feature extraction, and a hybrid cross-attention fusion network for multimodal feature integration. A lightweight student model is constructed by reducing the hidden feature dimension, the number of attention heads, and the feedforward network size while preserving the overall network architecture. The student model is trained using knowledge distillation to effectively transfer discriminative knowledge from the teacher. Finally, dynamic INT8 post-training quantization is applied to further reduce the model size for efficient deployment. Experimental results on the Audio-Visual Event (AVE) dataset show that the proposed framework reduces the number of trainable parameters in the multimodal fusion module by 59.06%, with only a 2.14% decrease in classification accuracy compared with the teacher model. Furthermore, dynamic INT8 quantization reduces the model size from 10.71 MB to 2.04 MB while maintaining competitive recognition performance. These results demonstrate that the proposed framework provides an effective trade-off between recognition accuracy and computational efficiency, making it a promising solution for deployment on resource-constrained edge AI platforms.

eess.SP↗

arXiv2026

Stable Hybrid Cross-Attention Fusion for Audio-Visual Event Recognition

Audio-Visual Event Recognition (AVER) is essential for intelligent urban monitoring systems, where robust multimodal understanding of complex environments is required. This paper proposes a stable hybrid cross-attention fusion framework for audio-visual event recognition in smart urban environments. The proposed architecture combines pretrained Video Masked Autoencoder (VideoMAE) and Audio Spectrogram Transformer (AST) representations with FiLM-based audio conditioning, bidirectional cross-attention fusion, multimodal Transformer encoding, and modality-temporal attention. To improve computational efficiency and training stability, frozen pretrained backbones and cached feature extraction are employed. Extensive experiments on the AVE dataset show that the proposed framework achieves the highest average performance among the evaluated unimodal and multimodal baselines across multiple evaluation metrics, obtaining a best validation accuracy of 91.74% and a test accuracy of 83.85 plus/minus 1.40% over five independent runs. The results indicate that the proposed hybrid fusion strategy effectively captures complementary audio-visual information and provides robust multimodal representation learning for challenging realworld urban monitoring scenarios.

eess.AS↗

arXiv2025

Investigation of Feature Selection and Pooling Methods for Environmental Sound Classification

This paper explores the impact of dimensionality reduction and pooling methods for Environmental Sound Classification (ESC) using lightweight CNNs. We evaluate Sparse Salient Region Pooling (SSRP) and its variants, SSRP-Basic (SSRP-B) and SSRP-Top-K (SSRP-T), under various hyperparameter settings and compare them with Principal Component Analysis (PCA). Experiments on the ESC-50 dataset demonstrate that SSRP-T achieves up to 80.69 % accuracy, significantly outperforming both the baseline CNN (66.75 %) and the PCA-reduced model (37.60 %). Our findings confirm that a well-tuned sparse pooling strategy provides a robust, efficient, and high-performing solution for ESC tasks, particularly in resource-constrained scenarios where balancing accuracy and computational cost is crucial.

eess.SP↗