MBW: Multi-view Bootstrapping in the Wild

TL;DR

This paper introduces MBW, a method that uses uncalibrated, handheld cameras and a neural prior to accurately estimate 2D and 3D landmarks of articulated objects with minimal supervision, enabling scalable analysis in natural settings.

Contribution

MBW combines a non-rigid 3D neural prior with deep flow to achieve high-fidelity landmark estimation from uncalibrated videos with only a few annotations, advancing beyond rigid, calibrated camera methods.

Findings

Achieves 2D landmark accuracy comparable to fully supervised methods with only 1-2% labeled frames.

Provides 3D reconstructions of articulated objects in natural, unstructured environments.

Demonstrates versatility across diverse species and objects in casual zoo videos.

Abstract

Labeling articulated objects in unconstrained settings have a wide variety of applications including entertainment, neuroscience, psychology, ethology, and many fields of medicine. Large offline labeled datasets do not exist for all but the most common articulated object categories (e.g., humans). Hand labeling these landmarks within a video sequence is a laborious task. Learned landmark detectors can help, but can be error-prone when trained from only a few examples. Multi-camera systems that train fine-grained detectors have shown significant promise in detecting such errors, allowing for self-supervised solutions that only need a small percentage of the video sequence to be hand-labeled. The approach, however, is based on calibrated cameras and rigid geometry, making it expensive, difficult to manage, and impractical in real-world scenarios. In this paper, we address these bottlenecks by combining a non-rigid 3D neural prior with deep flow to obtain high-fidelity landmark estimates from videos with only two or three uncalibrated, handheld cameras. With just a few annotations (representing 1-2% of the frames), we are able to produce 2D results comparable to state-of-the-art fully supervised methods, along with 3D reconstructions that are impossible with other existing approaches. Our Multi-view Bootstrapping in the Wild (MBW) approach demonstrates impressive results on standard human datasets, as well as tigers, cheetahs, fish, colobus monkeys, chimpanzees, and flamingos from videos captured casually in a zoo. We release the codebase for MBW as well as this challenging zoo dataset consisting image frames of tail-end distribution categories with their corresponding 2D, 3D labels generated from minimal human intervention.