Guochen Xie

TL;DR: The 3rd NTIRE challenge on single-image super-resolution (restoration of rich details in a low-resolution image) is reviewed with a focus on proposed solutions and results and the state-of-the-art in real-world single image super- resolution.

TL;DR: The multi-fold ensemble method is introduced to train and ensemble several models with the same architecture but different data distributions to enhance the performance of the solution based on the ensemble learning method.

TL;DR: An efficient pipeline based on optical flow features to spot micro expression (ME) and macro expression (MaE) spotting task and achieves the first place in the MEGC2022 Challenge.

TL;DR: This paper proposes an end-to-end unsupervised motion transfer network to tackle micro-expression generation and introduces face parsing method to pay specific attention to the eyeglasses regions to ensure the reasonability of the deformation.

TL;DR: In this article , a dual-branch based multi-output regression model was proposed to estimate emotional reaction intensity (ERI) in multimodal scenarios, and the spatial attention was used to better extract visual features and the Mel-Frequency Cepstral Coefficients technology extracted acoustic features.