Uncertainty modeling for fine-tuned implicit functions

TL;DR

This paper introduces Dropsembles, a new uncertainty estimation method for fine-tuned implicit functions like NeRFs and SDFs, enhancing reconstruction reliability especially under data sparsity and corruption.

Contribution

Dropsembles provides an efficient uncertainty estimation technique that matches deep ensembles in accuracy and calibration but requires less computational resources.

Findings

Dropsembles achieves comparable accuracy to deep ensembles.

Dropsembles offers better calibration than other methods.

Effective in real-world medical imaging scenarios.

Abstract

Implicit functions such as Neural Radiance Fields (NeRFs), occupancy networks, and signed distance functions (SDFs) have become pivotal in computer vision for reconstructing detailed object shapes from sparse views. Achieving optimal performance with these models can be challenging due to the extreme sparsity of inputs and distribution shifts induced by data corruptions. To this end, large, noise-free synthetic datasets can serve as shape priors to help models fill in gaps, but the resulting reconstructions must be approached with caution. Uncertainty estimation is crucial for assessing the quality of these reconstructions, particularly in identifying areas where the model is uncertain about the parts it has inferred from the prior. In this paper, we introduce Dropsembles, a novel method for uncertainty estimation in tuned implicit functions. We demonstrate the efficacy of our approach through a series of experiments, starting with toy examples and progressing to a real-world scenario. Specifically, we train a Convolutional Occupancy Network on synthetic anatomical data and test it on low-resolution MRI segmentations of the lumbar spine. Our results show that Dropsembles achieve the accuracy and calibration levels of deep ensembles but with significantly less computational cost.