Progressive Meta-Pooling Learning for Lightweight Image Classification Model
This addresses the challenge of improving accuracy for lightweight models on edge devices, though it appears incremental as it builds on existing MobileNetV2 architecture.
The paper tackles the problem of restricted receptive fields in lightweight image classification models for edge devices by proposing a Meta-Pooling framework with a Progressive Meta-Pooling Learning strategy, achieving a 2.3% improvement in top1 accuracy on ImageNet for MobileNetV2.
Practical networks for edge devices adopt shallow depth and small convolutional kernels to save memory and computational cost, which leads to a restricted receptive field. Conventional efficient learning methods focus on lightweight convolution designs, ignoring the role of the receptive field in neural network design. In this paper, we propose the Meta-Pooling framework to make the receptive field learnable for a lightweight network, which consists of parameterized pooling-based operations. Specifically, we introduce a parameterized spatial enhancer, which is composed of pooling operations to provide versatile receptive fields for each layer of a lightweight model. Then, we present a Progressive Meta-Pooling Learning (PMPL) strategy for the parameterized spatial enhancer to acquire a suitable receptive field size. The results on the ImageNet dataset demonstrate that MobileNetV2 using Meta-Pooling achieves top1 accuracy of 74.6\%, which outperforms MobileNetV2 by 2.3\%.