Kobaayyy/Awesome-ECCV2026-ECCV2024-ECCV2020-Low-Level-Vision

A Collection of Papers and Codes for ECCV2026/ECCV2024/ECCV2020 Low Level Vision

What it solves

Researchers and practitioners in computer vision often struggle to keep track of the many papers and code releases focused on low-level vision (tasks that modify or restore images at the pixel level). This repository solves that by gathering papers and code from three specific ECCV conference years (2026, 2024, and 2020) into one organized list, saving time and making it easier to find relevant work.

How it works

The repository is structured as a set of markdown files, each dedicated to a different ECCV year (e.g., ECCV2026.md, ECCV2024.md, ECCV2020.md). These files list papers along with links to their code when available. The collection covers a wide range of low-level vision subtopics, including super-resolution, image deraining, dehazing, deblurring, denoising, image restoration, image enhancement, demoireing, inpainting, image quality assessment, frame interpolation, and image/video compression. The project also links to related collections for other conferences (CVPR, ICCV) and to a community platform for low-level vision.

Who it's for

This list is aimed at researchers, students, and developers working in computer vision, particularly those interested in low-level image processing tasks. It helps them quickly locate state-of-the-art papers and their implementations from recent ECCV editions, and it encourages community contributions via star, fork, and pull requests.

Highlights

  • Covers ECCV 2026, 2024, and 2020 – three different years in one place
  • Tasks include super-resolution, deraining, dehazing, deblurring, denoising, restoration, enhancement, demoireing, inpainting, quality assessment, frame interpolation, and compression
  • Provides direct links to paper pages and code repositories when available
  • Encourages open contribution (star, fork, PR) and lists related collections for other major vision conferences

Related

  • Project
    Intellindust-AI-Lab/EdgeCrafter

    EdgeCrafter is a family of compact Vision Transformers for edge dense prediction tasks, including object detection, instance segmentation, and pose estimation. Using task-specialized distillation, it achieves high accuracy with small model sizes and low latency, ready for edge deployment on devices like Intel GPUs.

  • Project
    LogosIntelligenceLab/SeaFormer

    SeaFormer is a lightweight transformer architecture designed to provide high-performance image classification and semantic segmentation specifically for mobile devices.

  • Project
    xevrion/gyotaku

    A local, privacy-focused tool that uses OCR to index the text inside your screenshots, allowing you to search and retrieve them instantly by content.

  • Project
    RapidAI/RapidOcrOnnx

    A cross-platform OCR implementation using ONNX Runtime for efficient text detection, classification, and recognition across Windows, macOS, and Linux.