/Incognito

"Ctrl + Alt + Delete" your background

Primary LanguagePythonMIT LicenseMIT

Incognito

This project removes your background by optimizing segmentation models for CPU inference. It applies deep learning optimizations & engineering optimizations to improve CPU inference frame rate from 3 FPS to 15 FPS.

Results

Visual Results

        Original            Unet with ResNet        Unet with MobileNet

 UNet-EfficientNet EB0     UNet-EfficientNet EB1     UNet-EfficientNet EB2

Optimization Results

Optimization CPU Inference time FPS Model Size IOU
ResNet Model (256*256) ~265 ms 3-4 98.2 MB 0.95592
Architecture Optimization (EfficientNet EB0) ~229 ms 5-6 41.2 MB 0.96079
Deep Learning Optimization (Quantization + Graph Optimization + EfficientNet EB0) ~219 ms 5-6 10.6 MB 0.92648
Engineering Optimization (Quantization + Graph Optimization + EfficientNet EB0 + Image resolution reduction - 128*128) ~56 ms 15-17 10.6 MB 0.92648

Model files

The optimized and compressed models can be found under following directory

/data/assets/saved_models

Requirements (to-do)

Add docker image

Presentation Slides

Further details regarding the motivation, methods and results of different optimization techniques can be found in my presentation here.

License

MIT License

Copyright (c) 2019 Chinmay Naik