Quantization Effects on Neural Networks Perception: How would quantization change the perceptual field of vision models? (2403.09939v2)
Abstract: Neural network quantization is a critical technique for deploying models on resource-limited devices. Despite its widespread use, the impact of quantization on model perceptual fields, particularly in relation to class activation maps (CAMs), remains underexplored. This study investigates how quantization influences the spatial recognition abilities of vision models by examining the alignment between CAMs and visual salient objects maps across various architectures. Utilizing a dataset of 10,000 images from ImageNet, we conduct a comprehensive evaluation of six diverse CNN architectures: VGG16, ResNet50, EfficientNet, MobileNet, SqueezeNet, and DenseNet. Through the systematic application of quantization techniques, we identify subtle changes in CAMs and their alignment with Salient object maps. Our results demonstrate the differing sensitivities of these architectures to quantization and highlight its implications for model performance and interpretability in real-world applications. This work primarily contributes to a deeper understanding of neural network quantization, offering insights essential for deploying efficient and interpretable models in practical settings.
- “A survey of quantization methods for efficient neural network inference,” in Low-Power Computer Vision, pp. 291–326. Chapman and Hall/CRC, 2022.
- “Quantized convolutional neural networks for mobile devices,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 4820–4828.
- “Resource-efficient neural networks for embedded systems,” arXiv preprint arXiv:2001.03048, 2020.
- “Deploy large-scale deep neural networks in resource constrained iot devices with local quantization region,” arXiv preprint arXiv:1805.09473, 2018.
- “Learning deep features for discriminative localization,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2921–2929.
- “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 618–626.
- “Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks,” in 2018 IEEE winter conference on applications of computer vision (WACV). IEEE, 2018, pp. 839–847.
- Shifts in Selective Visual Attention: Towards the Underlying Neural Circuitry, pp. 115–141, Springer Netherlands, Dordrecht, 1987.
- “ImageNet: A Large-Scale Hierarchical Image Database,” in CVPR09, 2009.
- “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014.
- “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
- “Efficientnet: Rethinking model scaling for convolutional neural networks,” in International conference on machine learning. PMLR, 2019, pp. 6105–6114.
- “Mobilenetv2: Inverted residuals and linear bottlenecks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 4510–4520.
- “Squeezenet: Alexnet-level accuracy with 50x fewer parameters and¡ 0.5 mb model size,” arXiv preprint arXiv:1602.07360, 2016.
- “Densely connected convolutional networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 4700–4708.
- “Satsal: A multi-level self-attention based architecture for visual saliency prediction,” IEEE Access, vol. 10, pp. 20701–20713, 2022.
- “What do different evaluation metrics tell us about saliency models?,” IEEE transactions on pattern analysis and machine intelligence, vol. 41, no. 3, pp. 740–757, 2018.
- “The earth mover’s distance as a metric for image retrieval,” International Journal of Computer Vision, vol. 40, pp. 99–121, 2000.