Parallel Algorithm of IDCT with GPUs and CUDA for Large-scale Video Quality of 3G

When video is transmitted over 3G networks, the video quality might suffer from impairments caused by packet losses. Extracting video quality features is a set of algorithms and inverse discrete cosine transforms is an important algorithm in this field. To improve the performance and be suitable to apply to evaluating the 3G video quality in real-time, two different parallel algorithms with CUDA of the inverse discrete cosine transform are proposed. The parallel algorithms are exploited combined the high parallelism of GPUs with a high bandwidth in memory transfer at low cost. Moreover, the device memory accessing model is analyzed and considered the optimal use of the data cache. The experimental results show that the uberkernel algorithm with shared memory and texture memory outperforms other schemes for medium-sized tasks and the perisistent algorithm is better for large-scale tasks.