Towards Metamerism via Foveated Style Transfer
Authors: Arturo Deza, Aditya Jonnalagadda, Miguel P. Eckstein
ICLR 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our contributions include: ... 3) An ABX psychophysical evaluation of our metamers where we also find that the rate of growth of the receptive fields in our model match V1 for reference metamers and V2 between synthesized samples. Our model also renders metamers at roughly a second, presenting a 1000 speed-up compared to the previous work, which allows for tractable data-driven metamer experiments. ... Section 4 Experiments The goal of Experiment 1 is to estimate γ as a function of s via a computational simulation as a proxy for running human psychophysics. Once it is computed, we have reduced our minimization to a tractable single variable optimization problem. We will then proceed to Experiment 2 where we will perform an ABX experiment on human observers by varying the scale to render visual metamers... |
| Researcher Affiliation | Academia | Arturo Deza1,4, Aditya Jonnalagadda3, Miguel P. Eckstein1,2,4 1 Dynamical Neuroscience, 2Psychological and Brain Sciences, 3Electric and Computer Engineering, 4 Institute for Collaborative Biotechnologies UC Santa Barbara, CA, USA |
| Pseudocode | Yes | Algorithm 1 Pipeline for Metamer hyperparameter γ( ) search |
| Open Source Code | No | The paper acknowledges the use of code from others ('Xun Huang for sharing his code', 'Jeremy Freeman for making his metamer code available') but does not state that its own source code for the described methodology is publicly available. |
| Open Datasets | Yes | We used the publicly available code by Huang and Belongie for our decoder which was trained on Image Net and a collection of publicly available paintings to learn how to invert texture as well. |
| Dataset Splits | No | The paper describes the training of its internal modules (decoder, pix2pix refinement) but does not provide specific train/validation/test dataset splits for the main experiments using the 10 images shown in Figure 6. |
| Hardware Specification | Yes | Our pix2pix U-Net refinement module took 3 days to train on a Titan X GPU, and was trained with 64 crops (256 256) per image on 100 images, including horizontally mirrored versions. ... We would also like to thank NVIDIA for donating a Titan X GPU. |
| Software Dependencies | No | The paper references specific architectures and models (VGG19, AdaIN, pix2pix U-Net) and external code used, but does not provide specific software version numbers (e.g., library versions, framework versions) for reproduction. |
| Experiment Setup | Yes | The dimensionality of the input of the encoder is 1 512 512, and the dimensionality of the output (relu4_1) is 512 64 64... Our pix2pix U-Net refinement module took 3 days to train on a Titan X GPU, and was trained with 64 crops (256 256) per image on 100 images, including horizontally mirrored versions. We ran 200 training epochs of these 12800 images on the U-Net architecture proposed by Isola et al. (2017) which preserves local image structure given an adversarial and L2 loss. ... Images were rendered at 512 512 px, and we fixed the monitor at 52cm viewing distance and 800 600px resolution so that the stimuli subtended 26deg 26deg. The monitor was linearly calibrated with a maximum luminance of 115.83 2.12 cd/m2. |