DeepAI AI Chat
Log In Sign Up

Probabilistic orientation estimation with matrix Fisher distributions

by   D. Mohlin, et al.
KTH Royal Institute of Technology

This paper focuses on estimating probability distributions over the set of 3D rotations (SO(3)) using deep neural networks. Learning to regress models to the set of rotations is inherently difficult due to differences in topology between R^N and SO(3). We overcome this issue by using a neural network to output the parameters for a matrix Fisher distribution since these parameters are homeomorphic to R^9. By using a negative log likelihood loss for this distribution we get a loss which is convex with respect to the network outputs. By optimizing this loss we improve state-of-the-art on several challenging applicable datasets, namely Pascal3D+, ModelNet10-SO(3) and UPNA head pose.


page 7

page 13

page 15

page 19


Probabilistic Regression with Huber Distributions

In this paper we describe a probabilistic method for estimating the posi...

On the Pitfalls of Heteroscedastic Uncertainty Estimation with Probabilistic Neural Networks

Capturing aleatoric uncertainty is a critical part of many machine learn...

On Closed-Form expressions for the Fisher-Rao Distance

The Fisher-Rao distance is the geodesic distance between probability dis...

Evidential Softmax for Sparse Multimodal Distributions in Deep Generative Models

Many applications of generative models rely on the marginalization of th...

On the loss of Fisher information in some multi-object tracking observation models

The concept of Fisher information can be useful even in cases where the ...

Uncertainty Estimation by Fisher Information-based Evidential Deep Learning

Uncertainty estimation is a key factor that makes deep learning reliable...

pseudo-Bayesian Neural Networks for detecting Out of Distribution Inputs

Conventional Bayesian Neural Networks (BNNs) are known to be capable of ...