On Bayesian Adaptive Video Super Resolution

By Ce Liu, Deqing Sun. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014

357 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

VRT: A Video Restoration Transformer (2022, methodology): “All models are trained on Vimeo-90K [31] for bicubic video SR ( × 4) and tested it on Vid4 [61].”
TDAN: Temporally-Deformable Alignment Network for Video Super-Resolution (2018, methodology): “We evaluated our models on the Vid4 benchmark [16], which contains four video sequences: city , walk , calendar , and foliage , as other methods.”
Bidirectional Recurrent Convolutional Networks for Multi-Frame Super-Resolution (2015, methodology): “We use 25 YUV format video sequences3 as our training set, which have been widely used in many video SR methods [30], [35], [42].”
Video Enhancement with Task-Oriented Flow (2017, background): “e for denoising/deblocking and superresolution is 10 4, and the learning rate for interpolation is 3 410 . 5. The Vimeo-90K Dataset To acquire high quality videos for video processing, previous works [17, 14] take videos by themselves, resulting in video datasets that are small in size and limited i”
Video Super-Resolution With Convolutional Neural Networks (2016, methodology): “The source code of the Bayesian adaptive video SR method from [3] (Bayesian) is unavailable, thus, we used the reimplementation provided and used by Ma et al. [5] in their paper.”