Advances in Image and Video Enhancement

The field of image and video enhancement is moving towards developing more efficient and effective methods for improving the quality of low-resolution or degraded visual content. Researchers are exploring innovative approaches that combine advanced techniques such as super-resolution, feature upsampling, and neural networks to achieve state-of-the-art results. Noteworthy papers in this area include Adaptive Object Detection with ESRGAN-Enhanced Resolution & Faster R-CNN, which proposes a method for improved object detection from low-resolution images. Another notable paper is MambaVSR, which introduces a content-aware scanning state space model for video super-resolution. Additionally, FADPNet and FGA-NN demonstrate promising results in face super-resolution and film grain analysis, respectively. MSNeRV also shows impressive performance in neural video representation with multi-scale feature fusion.

Sources

Adaptive Object Detection with ESRGAN-Enhanced Resolution & Faster R-CNN

JAFAR: Jack up Any Feature at Any Resolution

MambaVSR: Content-Aware Scanning State Space Model for Video Super-Resolution

FADPNet: Frequency-Aware Dual-Path Network for Face Super-Resolution

FGA-NN: Film Grain Analysis Neural Network

MSNeRV: Neural Video Representation with Multi-Scale Feature Fusion

Built with on top of