IMDLBenCo Documentation
Benchmark and Codebase for Image manipulation localization & detection
Modular Design
Based on object-oriented encapsulation, supplemented by a few callback functions and registration mechanisms, making it easy to call and redevelop.
Research-Oriented
A framework designed for the characteristics of Image manipulation detection/localization (IMDL) research. It includes rich IMDL preprocessing, highly customizable model and loss function design, and efficient GPU-accelerated evaluation metric computation.
User-Friendly
Installation via PyPI and command line invocation, separating training scripts from the core framework code, allowing you to customize your workflow without modifying the source code, making it easy to get started.
Important! The current documentation and tutorials are not complete. This is a project that requires a lot of manpower, and we will do our best to complete it as quickly as possible.
Currently, you can use the demo by following the brief tutorial provided in the Readme of the official Github repository!