Deep screen image crop and enhance Week 3 (Aaron Ott, Amir Mazaheri)
Problem We have taken a photo of an image, and we want the original image. This network for this can be broken into 2 parts: Image Detector/Cropper Image Enhancer
Cropper Uses a frozen VGG-19 model to get feature map Applies convolutions, normalizations, and activations Final dense layer creates 6-number theta value for affine transformation STN takes input image and applies affine transformation
Enhancer Pretrained EDSR (trained on DIV2K) Modified form of Resnet https://github.com/krasserm/super-resolution Pretrained EDSR (trained on DIV2K) Modified form of Resnet Uses modified residual block, which excludes batch normalization and final ReLU layer 16 Residual blocks Subpixel Conv2D layers for upscaling the image Scales the image 4x Lim, Son, Kim, Nah, Lee. “Enhanced Deep Residual Networks for Single Image Super-Resolution”. 10 July 2017
Combined Cropper and Enhancer Trained with 2 outputs and 2 Loss Functions: - Trained Cropper on VGG + Cosine Proximity (Inception Loss) - Trained Enhancer on VGG + MSE
Results Metric\Model Cropper Cropper & Enhancer PSNR 11.1903 16.2060 SSIM 0.4254 0.4909 MSE 0.0796 0.0281 MOS 2.6143 2.8857 Results Cropper & Enhancer Input Cropper Actual
Shortcomings of PSNR, SSIM, and MSE Metrics: PSNR: 18.3130 SSIM: 0.5358 MSE: 0.0164 Input Output Loss Functions VGG + Cosine Proximity MSE Actual
Synthetic Dataset Problem: There is no existing dataset to use when solving this problem, and taking pictures takes too much time Solution: Automatically generate images with various transformations over various backgrounds - Current problems: sometimes image edges get cut out, difficult to get full variety of possible images, doesn’t yet account for discoloration or image noise, dataset only includes birds http://www.vision.caltech.edu/visipedia/CUB-200.html, http://places2.csail.mit.edu/download.html
Synthetic Dataset Results Cropper + Enhancer Metric\Model Cropper + Enhancer Cropper PSNR 16.3862 17.2877 SSIM 0.4882 0.5421 MSE 0.0262 0.0212 Input Cropper Truth
What’s Next Short Term (next week) Train cropper on harder synthetic dataset Expand synthetic dataset to include multiple image classes Develop GAN to train model further Long Term (to the end of the summer) Connect model to solve existing issues: identification/classification Try new enhancers and enhancement techniques