1. Fixed several bugs in the face detector with multi-level input. 2. Update the contents of the benchmark report.
1.1.1
1. Verified the feasibility of compiling on the iOS platform under the action. 2. Added an image interface for I/O testing. 3. Updated and corrected some errors or omissions in the documentation.
1.1.0
1. Added tracking-by-detection tracking mode. 2. Adapted to multiple levels of detection modes. 3. Added support for CUDA Device accelerated inference. 4. Support script compilation for iOS framework.
1.0.3
Add a build toolchain based on ios-Cmake to build the iOS framework in macos with one click.
1.0.2
Added SDK compilation related scripts for Android version.
1. Replace the previous tripartite dependency library with submodule. 2. Add precompiled software for each architecture. 3. Add CI program for automatic testing in Actions. 4. Add automatic compilation of multi-platform SDKs in Actions. 5. Add Linux aarch64. 6. Write quick compilation scripts in docker.