v0.4.0

New Additions

• Added Android sample applications for Super Resolution and Image Classification under apps/Android.
• Added Models: whisper_small_en, whisper_base_en, whisper_tiny_en, aotgan
• Added quantized models: quicksrnetlarge_quantized, quicksrnetmedium_quantized, quicksrnetsmall_quantized, resnext50_quantized
• Add on-device demo for selfie segmentation

Quality Improvements & Bug Fixes

• Consolidated requirements where possible to eliminate thrashing when using multiple models in 1 environment
• Added "protocols" (models/protocols.py) for better type checking of base classes.
• Updates to various files in the models folder, working towards enablement of MyPy.
• googlenet_quantized and mobilenet_v3_large_quantized overhauled to improve quantization performance

Performance Numbers

• Added Galaxy S24 Performance Numbers
• Updated existing numbers to reflect benchmarks from latest AI Hub toolchain