Fast Fourier Intrinsic Network

  title={Fast Fourier Intrinsic Network},
  author={Yanlin Qian and Miaojing Shi and Joni-Kristian Kamarainen and Jiri Matas},
  journal={2021 IEEE Winter Conference on Applications of Computer Vision (WACV)},
We address the problem of decomposing an image into albedo and shading. We propose the Fast Fourier Intrinsic Network, FFI-Net in short, that operates in the spectral domain, splitting the input into several spectral bands. Weights in FFI-Net are optimized in the spectral domain, allowing faster convergence to a lower error. FFI-Net is lightweight and does not need auxiliary networks for training. The network is trained end-to-end with a novel spectral loss which measures the global distance… 


Intrinsic Image Transformation via Scale Space Decomposition
A new network structure for decomposing an image into its intrinsic albedo and shading is introduced, and a multi-channel architecture that learns the image-to-image transformation function in successive frequency bands in parallel is a fully convolutional neural network.
Self-Supervised Intrinsic Image Decomposition
This paper proposes a model that joins together an image decomposition pipeline, which predicts reflectance, shape, and lighting conditions given a single image, with a recombination function, a learned shading model used to recompose the original input based off of intrinsic image predictions, and a network that can use unsupervised reconstruction error as an additional signal to improve its intermediate representations.
Direct Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression
The strategy is to learn a convolutional neural network that directly predicts output albedo and shading channels from an input RGB image patch, which outperforms all prior work, including methods that rely on RGB+Depth input.
Revisiting Deep Intrinsic Image Decompositions
This work adopts core network structures that universally reflect loose prior knowledge regarding the intrinsic image formation process and can be largely shared across datasets, and applies flexibly supervised loss layers that are customized for each source of ground truth labels.
DARN: A Deep Adversarial Residual Network for Intrinsic Image Decomposition
Compared to recent deep learning techniques, this work simplifies the architecture, making it easier to build and train, and constrain it to generate a valid and reversible decomposition, and outperforms the state of the art deep algorithms both on the qualitative and quantitative aspect.
CGIntrinsics: Better Intrinsic Image Decomposition through Physically-Based Rendering
CGIntrinsics, a new, large-scale dataset of physically-based rendered images of scenes with full ground truth decompositions, is presented, demonstrating the suprising effectiveness of carefully-rendered synthetic data for the intrinsic images task.
Learning Non-Lambertian Object Intrinsics Across ShapeNet Categories
This work focuses on the non-Lambertian object-level intrinsic problem of recovering diffuse albedo, shading, and specular highlights from a single image of an object, and shows that feature learning at the encoder stage is more crucial for developing a universal representation across categories.
Learning Data-Driven Reflectance Priors for Intrinsic Image Decomposition
A model is trained to predict relative reflectance ordering between image patches from large-scale human annotations, producing a data-driven reflectance prior and it is shown how to naturally integrate this learned prior into existing energy minimization frame-works for intrinsic image decomposition.
Unsupervised Learning for Intrinsic Image Decomposition From a Single Image
This paper proposes a novel unsupervised intrinsic image decomposition framework, which relies on neither labeled training data nor hand-crafted priors, and directly learns the latent feature of reflectance and shading from unsuper supervised and uncorrelated data.
Single Image Intrinsic Decomposition Without a Single Intrinsic Image
This paper presents a two stream convolutional neural network framework that is capable of learning the decomposition effectively in the absence of any ground truth intrinsic images, and can be easily extended to a (semi-)supervised setup.