Amplitude Spectrum Transformation for Open Compound Domain Adaptive Semantic Segmentation

Authors

  • Jogendra Nath Kundu Indian Institute of Science
  • Akshay R Kulkarni Indian Institute of Science
  • Suvaansh Bhambri Indian Institute of Science
  • Varun Jampani Google
  • Venkatesh Babu Radhakrishnan Indian Institute of Science

DOI:

https://doi.org/10.1609/aaai.v36i2.20008

Keywords:

Computer Vision (CV), Machine Learning (ML), Domain(s) Of Application (APP)

Abstract

Open compound domain adaptation (OCDA) has emerged as a practical adaptation setting which considers a single labeled source domain against a compound of multi-modal unlabeled target data in order to generalize better on novel unseen domains. We hypothesize that an improved disentanglement of domain-related and task-related factors of dense intermediate layer features can greatly aid OCDA. Prior-arts attempt this indirectly by employing adversarial domain discriminators on the spatial CNN output. However, we find that latent features derived from the Fourier-based amplitude spectrum of deep CNN features hold a more tractable mapping with domain discrimination. Motivated by this, we propose a novel feature space Amplitude Spectrum Transformation (AST). During adaptation, we employ the AST auto-encoder for two purposes. First, carefully mined source-target instance pairs undergo a simulation of cross-domain feature stylization (AST-Sim) at a particular layer by altering the AST-latent. Second, AST operating at a later layer is tasked to normalize (AST-Norm) the domain content by fixing its latent to a mean prototype. Our simplified adaptation technique is not only clustering-free but also free from complex adversarial alignment. We achieve leading performance against the prior arts on the OCDA scene segmentation benchmarks.

Downloads

Published

2022-06-28

How to Cite

Kundu, J. N., Kulkarni, A. R., Bhambri, S., Jampani, V., & Radhakrishnan, V. B. (2022). Amplitude Spectrum Transformation for Open Compound Domain Adaptive Semantic Segmentation. Proceedings of the AAAI Conference on Artificial Intelligence, 36(2), 1220-1227. https://doi.org/10.1609/aaai.v36i2.20008

Issue

Section

AAAI Technical Track on Computer Vision II