TY - GEN
T1 - Fast fourier intrinsic network
AU - Qian, Yanlin
AU - Shi, Miaojing
AU - Kämäräinen, Joni-Kristian
AU - Matas, Jiri
N1 - Funding Information:
Acknowledgement: This work was partially supported by the National Natural Science Foundation of China (NSFC) under Grant No. 61828602.
Publisher Copyright:
© 2021 IEEE.
jufoid=57596
PY - 2021/1
Y1 - 2021/1
N2 - 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 between the network prediction and corresponding ground truth. FFI-Net achieves state-of-the-art performance on MPI-Sintel, MIT Intrinsic, and IIW datasets.
AB - 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 between the network prediction and corresponding ground truth. FFI-Net achieves state-of-the-art performance on MPI-Sintel, MIT Intrinsic, and IIW datasets.
U2 - 10.1109/WACV48630.2021.00321
DO - 10.1109/WACV48630.2021.00321
M3 - Conference contribution
AN - SCOPUS:85116140635
T3 - Proceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021
SP - 3168
EP - 3177
BT - Proceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021
PB - IEEE
T2 - IEEE Winter Conference on Applications of Computer Vision
Y2 - 5 January 2021 through 9 January 2021
ER -