Learning how to be robust: Deep polynomial regression

Abstract : Polynomial regression is a recurrent problem with a large number of applications. In computer vision it often appears in motion analysis. Whatever the application, standard methods for regression of polynomial models tend to deliver biased results when the input data is heavily contaminated by outliers. Moreover, the problem is even harder when outliers have strong structure. Departing from problem-tailored heuristics for robust estimation of parametric models, we explore deep convolutional neural networks. Our work aims to find a generic approach for training deep regression models without the explicit need of supervised annotation. We bypass the need for a tailored loss function on the regression parameters by attaching to our model a differentiable hard-wired decoder corresponding to the polynomial operation at hand. We demonstrate the value of our findings by comparing with standard robust regression methods. Furthermore, we demonstrate how to use such models for a real computer vision problem, i.e., video stabilization. The qualitative and quantitative experiments show that neural networks are able to learn robustness for general polynomial regression, with results that well overpass scores of traditional robust estimation methods.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [42 references]  Display  Hide  Download

https://hal.inria.fr/hal-01923068
Contributor : Juan-Manuel Perez-Rua <>
Submitted on : Wednesday, November 14, 2018 - 9:19:31 PM
Last modification on : Friday, April 19, 2019 - 4:55:05 PM
Long-term archiving on : Friday, February 15, 2019 - 5:11:23 PM

File

1804.06504.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01923068, version 1

Collections

Citation

Juan-Manuel Pérez-Rúa, Tomas Crivelli, Patrick Bouthemy, Patrick Pérez. Learning how to be robust: Deep polynomial regression. 2018. ⟨hal-01923068⟩

Share

Metrics

Record views

58

Files downloads

405