BACON: Band-limited Coordinate Networks for Multiscale Scene Representation

12/09/2021
by   David B. Lindell, et al.
0

Coordinate-based networks have emerged as a powerful tool for 3D representation and scene reconstruction. These networks are trained to map continuous input coordinates to the value of a signal at each point. Still, current architectures are black boxes: their spectral characteristics cannot be easily analyzed, and their behavior at unsupervised points is difficult to predict. Moreover, these networks are typically trained to represent a signal at a single scale, so naive downsampling or upsampling results in artifacts. We introduce band-limited coordinate networks (BACON), a network architecture with an analytical Fourier spectrum. BACON has constrained behavior at unsupervised points, can be designed based on the spectral characteristics of the represented signal, and can represent signals at multiple scales without per-scale supervision. We demonstrate BACON for multiscale neural representation of images, radiance fields, and 3D scenes using signed distance functions and show that it outperforms conventional single-scale coordinate networks in terms of interpretability and quality.

READ FULL TEXT

page 9

page 10

page 21

page 24

page 28

page 35

page 36

page 38

research
05/06/2021

ACORN: Adaptive Coordinate Networks for Neural Scene Representation

Neural representations have emerged as a new paradigm for applications i...
research
06/01/2022

Residual Multiplicative Filter Networks for Multiscale Reconstruction

Coordinate networks like Multiplicative Filter Networks (MFNs) and BACON...
research
02/07/2022

MINER: Multiscale Implicit Neural Representations

We introduce a new neural signal representation designed for the efficie...
research
10/20/2022

Coordinates Are NOT Lonely – Codebook Prior Helps Implicit Neural 3D Representations

Implicit neural 3D representation has achieved impressive results in sur...
research
05/15/2023

Curvature-Aware Training for Coordinate Networks

Coordinate networks are widely used in computer vision due to their abil...
research
11/29/2022

HashEncoding: Autoencoding with Multiscale Coordinate Hashing

We present HashEncoding, a novel autoencoding architecture that leverage...

Please sign up or login with your details

Forgot password? Click here to reset