Gaussian Process Priors for View-Aware Inference
Preprint
Abstract
While frame-independent predictions with deep neural networks have become the prominent solutions to many computer vision tasks, the potential benefits of utilizing correlations between frames have received less attention. Even though probabilistic machine learning provides the ability to encode correlation as prior knowledge for inference, there is a tangible gap between the theory and practice of applying probabilistic methods to modern vision problems. For this, we derive a principled framework to combine information coupling between camera poses (translation and orientation) with deep models. We proposed a novel view kernel that generalizes the standard periodic kernel in $\mathrm{SO}(3)$. We show how this soft-prior knowledge can aid several pose-related vision tasks like novel view synthesis and predict arbitrary points in the latent space of generative models, pointing towards a range of new applications for inter-frame reasoning.
Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Gaussian process for 6-DoF rigid motionsAutonomous Robots, 2017
- Computationally Efficient Rigid-Body Gaussian Process for Motion DynamicsIEEE Robotics and Automation Letters, 2017
- Probabilistic numerics and uncertainty in computationsProceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2015
- Spatiotemporal Learning via Infinite-Dimensional Bayesian Filtering and Smoothing: A Look at Gaussian Process Regression Through Kalman FilteringIEEE Signal Processing Magazine, 2013
- Rotation Invariant Kernels and Their Application to Shape AnalysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Rigid Body Dynamics Algorithms2008
- Invariant kernel functions for pattern analysis and machine learningMachine Learning, 2007
- Multiple View Geometry in Computer Vision2004