No abstract

Both efficient neural networks and hardware accelerators are being explored to speed up DNN inference on edge devices. For example, MobileNet uses depthwise separable convolution to achieve much lower latency, while systolic arrays provide much higher performance per watt. Interestingly however, the combination of these two ideas is inefficient: The computational patterns of depth-wise separable convolution are not systolic and lack data reuse to saturate the systolic array's constrained dataflow. In this paper, we propose FuSeConv (Fully-Separable Convolution) as a drop-in replacement for depth-wise separable convolution. FuSeConv generalizes the decomposition of convolutions fully to separable 1D convolutions along spatial and depth dimensions. The resultant computation is systolic and efficiently utilizes the systolic array with a slightly modified dataflow. With FuSeConv, we achieve a significant speed-up of 3x-7x with the MobileNet family of networks on a systolic array of size 64x64, with comparable accuracy on the ImageNet dataset. The high speed-up motivates exploration of hardware-aware Neural Operator Search (NOS) in complement to ongoing efforts on Neural Architecture Search (NAS).

No abstract

No abstract

In recent days, the data are transformed in the form of multimedia data such as images, graphics, audio and video. Multimedia data require a huge amount of storage capacity and transmission bandwidth. Consequently, data compression is used for reducing the data redundancy and serves more storage of data. In this paper, addresses the problem (demerits) of the lossy compression of images. This proposed method is deals on SVD Power Method that overcomes the demerits of Python SVD function. In our experimental result shows superiority of proposed compression method over those of Python SVD function and some various compression techniques. In addition, the proposed method also provides different degrees of error flexibility, which give minimum of execution of time and a better image compression.