PixelNN: Example-based Image Synthesis - The Robotics Institute Carnegie Mellon University
Home/PixelNN: Example-based Image Synthesis

PixelNN: Example-based Image Synthesis

Conference Paper, Proceedings of (ICLR) International Conference on Learning Representations, April, 2018
View Publication


We present a simple nearest-neighbor (NN) approach that synthesizes high-frequency photorealistic images from an "incomplete" signal such as a low-resolution image, a surface normal map, or edges. Current state-of-the-art deep generative models designed for such conditional image synthesis lack two important things: (1) they are unable to generate a large set of diverse outputs, due to the mode collapse problem. (2) they are not interpretable, making it difficult to control the synthesized output. We demonstrate that NN approaches potentially address such limitations, but suffer in accuracy on small datasets. We design a simple pipeline that combines the best of both worlds: the first stage uses a convolutional neural network (CNN) to maps the input to a (overly-smoothed) image, and the second stage uses a pixel-wise nearest neighbor method to map the smoothed output to multiple high-quality, high-frequency outputs in a controllable manner. We demonstrate our approach for various input modalities, and for various domains ranging from human faces to cats-and-dogs to shoes and handbags.


author = {A. Bansal and Y. Sheikh and D. Ramanan},
title = {PixelNN: Example-based Image Synthesis},
booktitle = {Proceedings of (ICLR) International Conference on Learning Representations},
year = {2018},
month = {April},

Share This Story!