Depth Contrast: Self-Supervised Pretraining on 3DPM Images for Mining Material Classification

TL;DR

This paper introduces Depth Contrast, a self-supervised learning method using depth maps and RGB images from 3DPM sensors to classify mining materials without extensive human annotations, outperforming transfer learning methods.

Contribution

The paper proposes a novel self-supervised approach, Depth Contrast, specifically designed for mining material classification using sensor data, leveraging depth maps and transfer learning.

Findings

Outperforms ImageNet transfer learning in supervised settings.

Achieves 11% improvement over transfer learning with limited labels.

Shows better generalization in linear evaluation.

Abstract

This work presents a novel self-supervised representation learning method to learn efficient representations without labels on images from a 3DPM sensor (3-Dimensional Particle Measurement; estimates the particle size distribution of material) utilizing RGB images and depth maps of mining material on the conveyor belt. Human annotations for material categories on sensor-generated data are scarce and cost-intensive. Currently, representation learning without human annotations remains unexplored for mining materials and does not leverage on utilization of sensor-generated data. The proposed method, Depth Contrast, enables self-supervised learning of representations without labels on the 3DPM dataset by exploiting depth maps and inductive transfer. The proposed method outperforms material classification over ImageNet transfer learning performance in fully supervised learning settings and achieves an F1 score of 0.73. Further, The proposed method yields an F1 score of 0.65 with an 11% improvement over ImageNet transfer learning performance in a semi-supervised setting when only 20% of labels are used in fine-tuning. Finally, the Proposed method showcases improved performance generalization on linear evaluation. The implementation of proposed method is available on GitHub.