CC-Net: Image Complexity Guided Network Compression for Biomedical Image Segmentation
This addresses the problem of high memory and latency in biomedical image analysis for researchers and practitioners, but it is incremental as it builds on existing compression methods.
The paper tackles the challenge of compressing large convolutional neural networks for biomedical image segmentation by proposing CC-Net, which uses image complexity to predict accuracy and select compression factors, resulting in compressed networks that retain up to 95% accuracy with only 0.1% of parameters.
Convolutional neural networks (CNNs) for biomedical image analysis are often of very large size, resulting in high memory requirement and high latency of operations. Searching for an acceptable compressed representation of the base CNN for a specific imaging application typically involves a series of time-consuming training/validation experiments to achieve a good compromise between network size and accuracy. To address this challenge, we propose CC-Net, a new image complexity-guided CNN compression scheme for biomedical image segmentation. Given a CNN model, CC-Net predicts the final accuracy of networks of different sizes based on the average image complexity computed from the training data. It then selects a multiplicative factor for producing a desired network with acceptable network accuracy and size. Experiments show that CC-Net is effective for generating compressed segmentation networks, retaining up to 95% of the base network segmentation accuracy and utilizing only 0.1% of trainable parameters of the full-sized networks in the best case.