Wavelet-SRNet: A Wavelet-Based CNN for Multi-scale Face Super Resolution

By Huaibo Huang, Ran He, Zhenan Sun, Tien-Ping Tan. IEEE International Conference on Computer Vision, 2017

520 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/708f4787bec9d7563f4bb8b33834de445147133b

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues (2020, methodology): “Frequency domain analysis is a classical and important method in image signal processing and has been widely used in a number of applications such as image classification [53, 52, 23], steganalysis [17, 10], texture classification [28, 25, 22] and super-resolution [39, 31].”
To learn image super-resolution, use a GAN to learn how to do image degradation first (2018, methodology): “SRGAN and Wavelet-SRNet were trained on our training set using pairs of bilinearly downsampled - HR images.”
HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment (2020, unknown): “719 WaveletCNN [14] 28.”
Blind Face Restoration via Deep Multi-scale Component Dictionaries (2020, background): “H and I. Input *RCAN *ESRGAN *WaveletSR GFRNet Ours Ground-truth Input *RCAN *WaveletSR GFRNet GWAINet Ours Ground-truth Input *RCAN *ESRGAN *WaveletSR GFRNet Ours Ground-truth Input *RCAN *WaveletSR GFRNet GWAINet Ours Ground-truth Blind Face Restoration via Deep Multi-scale Component Dictionaries ”
Latte: Latent Diffusion Transformer for Video Generation (2024, unknown): “…and got remarkable success in many fields, such as image inpainting (Ma et al, 2022, 2021, 2023), image super-resolution Luo et al (2022); Huang et al (2017), image cropping (Jia et al, 2022), forgery detection (Jia et al, 2021), face recognition (Luo et al, 2021a,b), nat-ural language…”
FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors (2017, methodology): “We next compare FSRGAN with two recent face SR methods: Wavelet-SRNet [12] and CBN [46].”
Wavelet Convolutions for Large Receptive Fields (2024, methodology): “[30] and Guo et al .”
Super-FAN: Integrated Facial Landmark Localization and Super-Resolution of Real-World Low Resolution Faces in Arbitrary Poses with GANs (2017, methodology): “As opposed to prior work on face super-resolution which primarily uses frontal datasets [31, 5, 12, 32, 30, 28] (e.g. CelebA, Helen, LFW, BioID) to report results, the low resolution images used in our experiments were generated using the newly created LS3D-W balanced dataset [4] which contains an…”
Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation (2019, methodology): “One is to reconstruct HR query images from LR ones before recognition [13, 26, 28], which is called hallucination method.”
GAN Prior Embedded Network for Blind Face Restoration in the Wild (2021, background): “[17] presented a wavelet-based approach that can ultra-resolve a very low-resolution (LR) face image.”
Deep Learning for Image Super-Resolution: A Survey (2019, methodology): “Similarly, the DWSR [117] and Wavelet-SRNet [118] also perform SR in the wavelet domain but with more complicated structures.”
IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis (2018, background): “They have achieved remarkable success in various tasks, such as unconditional or conditional image synthesis [22, 27], image-to-image translation [25, 46], image restoration [5, 17] and speech synthesis [12].”