Dilated convolution is used to achieve wide receptive fields in computer vision algorithms such as image segmentation and denoising. Unlike the strided convolution, dilated convolution maintains the resolution of the output feature map same as the input feature map. Thus, the computational complexity can be increased to configure the convolutional neural network (CNN) architecture with the dilated convolutional layer. However, the complexity accordingly introduces additional computation delay and it is strongly required to have a proper way to lessen the computation delay of the dilated convolution. In this paper, we propose the dilated-Winograd convolution to reduce the computational complexity of the dilated convolution. By using the Winograd transform with a dilation rate, the number of pixels in the tile is effectively reduced. The proposed acceleration methods result in an average speedup of 2.043 and 1.456 with dilation rate of 2 and 4 compared to the state-of-the-art implementation.

In this paper, a novel method to accelerate the spatial resolution downscaling transcoding operation for high efficiency video coding (HEVC) is proposed. The proposed transcoder first extracts the information about the coding unit (CU) structure during the decoding process and analyzes it to construct a pseudo-quadtree of the target resolution. By utilizing the constructed pseudo-quadtree, the encoder process is accelerated by searching only the optimal depth in the quadtree structure. To evaluate the performance of the proposed transcoding method, its transcoding time and bitrate loss are measured and compared with those of the HEVC reference encoder and decoder. Also, a comparison with state-of-the-art fast CU size decision methods and spatial resolution downscaling transcoding methods for HEVC is presented. The proposed method achieves an average speedup of 1.65 with an average bitrate loss of 2.06%.

Since the computer vision technologies are used to extract features from the edges of images, the median filter is frequently used as a pre-processing operation to improve the quality of feature objects. However, the median filter algorithms still have the limitation on integrating with the mobile systems due to the large computation overhead. Since the median filter accounts for a large proportion of total computation, this paper proposes a separable median filter to resolve the limitation. The proposed separable median filter is implemented with the sorting network, and reuses the data sorted at neighbor pixels, reducing duplicated comparisons.