论文标题
适应有效CNN的自适应像素结构稀疏网络
Adaptive Pixel-wise Structured Sparse Network for Efficient CNNs
论文作者
论文摘要
为了加速Deep CNN模型，本文提出了一个新型的空间自适应框架，可以根据输入图像动态生成像素的稀疏性。稀疏方案是像素的精致，在统一重要性图下的区域自适应，这使其对硬件实现友好。进一步提出了一种稀疏控制方法，以启用对具有不同精度/延迟要求的应用程序的在线调整。稀疏模型适用于广泛的视觉任务。实验结果表明，使用RESNET-18和使用SRResnet的超级分辨率有效地提高了图像分类的计算效率。在图像分类任务上，我们的方法可以节省30％-70％的MAC，而TOP-1和前5个精度略有下降。在超级分辨率任务上，我们的方法可以减少90％以上的MAC，而PSNR和SSIM的降低约为0.1 dB，而0.01降低了0.01。还包括硬件验证。
To accelerate deep CNN models, this paper proposes a novel spatially adaptive framework that can dynamically generate pixel-wise sparsity according to the input image. The sparse scheme is pixel-wise refined, regional adaptive under a unified importance map, which makes it friendly to hardware implementation. A sparse controlling method is further presented to enable online adjustment for applications with different precision/latency requirements. The sparse model is applicable to a wide range of vision tasks. Experimental results show that this method efficiently improve the computing efficiency for both image classification using ResNet-18 and super resolution using SRResNet. On image classification task, our method can save 30%-70% MACs with a slightly drop in top-1 and top-5 accuracy. On super resolution task, our method can reduce more than 90% MACs while only causing around 0.1 dB and 0.01 decreasing in PSNR and SSIM. Hardware validation is also included.
