Download
Abstract

We present methods for conditional and residual coding in the context of scalable coding for humans and machines. Our focus is on optimizing the rate-distortion performance of the reconstruction task using the information available in the computer vision task. We include an information analysis of both approaches to provide baselines and also propose an entropy model suitable for conditional coding with increased modelling capacity and similar tractability as previous work. We apply these methods to image reconstruction, using, in one instance, representations created for semantic segmentation on the Cityscapes dataset, and in another instance, representations created for object detection on the COCO dataset. In both experiments, we obtain similar performance between the conditional and residual methods, with the resulting rate-distortion curves contained within our baselines.

Citation

Anderson de Andrade, Alon Harell, & Ivan Bajić. (2023). “Conditional and residual methods in scalable coding for humans and machines.” ICME Workshop on Coding for Machines.

@inproceedings{DBLP:conf/icmcs/AndradeHFB23,
  author       = {Anderson de Andrade and
                  Alon Harell and
                  Yalda Foroutan and
                  Ivan V. Bajic},
  title        = {Conditional and Residual Methods in Scalable Coding for Humans and
                  Machines},
  booktitle    = {{IEEE} International Conference on Multimedia and Expo Workshops,
                  {ICMEW} Workshops 2023, Brisbane, Australia, July 10-14, 2023},
  pages        = {194--199},
  publisher    = {{IEEE}},
  year         = {2023},
  url          = {https://doi.org/10.1109/ICMEW59549.2023.00040},
  doi          = {10.1109/ICMEW59549.2023.00040},
  timestamp    = {Mon, 05 Feb 2024 17:45:49 +0100},
  biburl       = {https://dblp.org/rec/conf/icmcs/AndradeHFB23.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}