High resolution image

By Wern-Joo Sohn, Gi-Jeong Gwon, Hyeng-Soo Kim, Sanjiv Neupane, Sung-Jin Cho, Jae-Hyung Lee, Hitoshi Yamamoto, Je-Yong Choi, Chang-Hyeon An, Youngkyun Lee and 3 more. 2015

234 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/52d6900732207ae96674c58dd32f607f64ec3f7b

How researchers see this paper

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

1. Chen Change Loy
2. Yu Qiao
3. Luc Van Gool
4. Xintao Wang
5. Jinjin Gu
6. Xiaogang Wang
7. Limin Wang
8. Xiaoou Tang

Citations in the corpus

• BBDM: Image-to-Image Translation with Brownian Bridge Diffusion Models (2022, methodology): “In addition to GAN-based models, diffusion models [31] have also achieved impressive results on image generation [6, 12], inpainting [29], super-resolution [29, 30], and text-to-image generation [28].”
• Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators (2023, methodology): “We start this section with a brief introduction of diffusion models, particularly Stable Diffusion (SD) [29].”
• Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild (2024, methodology): “Since the low-frequency information of the image is mainly generated in the early stage of diffusion prediction [65] (where t and σ t are relatively large, and the weight k = ( σ t /σ T ) τ r is also large), the prediction result is closer to z LQ to enhance fidelity.”
• Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation (2021, background): “The problem is formulated in an image-to-image translation manner [27, 58], where massively-available talking face videos provide nat-ural self-supervision.”
• HDR image reconstruction from a single exposure using deep CNNs (2017, unknown): “Recent work on inpainting [Pathak et al. 2016; Yang et al. 2016] have also utilized variants of Generative Adversarial Networks (GANs) [Goodfellow et al.”
• VBench: Comprehensive Benchmark Suite for Video Generative Models (2023, methodology): “We choose 10 VBench dimensions that can encompass frame-wise generation capabilities, and sample frames from all the image and video generation models according to Prompt Suite per Evaluation Dimension described in Section 3.2.”