Improving 3D Semi-supervised Learning by Effectively Utilizing All Unlabelled Data

TL;DR

AllMatch is a novel SSL-based 3D classification framework that effectively utilizes all the unlabelled samples and exhibits its efficiency in effectively leveraging all the unlabelled data, demonstrated by the fact that only 10% of labelled data reaches nearly the same performance as fully-supervised learning with all labelled data.

Abstract

Semi-supervised learning (SSL) has shown its effectiveness in learning effective 3D representation from a small amount of labelled data while utilizing large unlabelled data. Traditional semi-supervised approaches rely on the fundamental concept of predicting pseudo-labels for unlabelled data and incorporating them into the learning process. However, we identify that the existing methods do not fully utilize all the unlabelled samples and consequently limit their potential performance. To address this issue, we propose AllMatch, a novel SSL-based 3D classification framework that effectively utilizes all the unlabelled samples. AllMatch comprises three modules: (1) an adaptive hard augmentation module that applies relatively hard augmentations to the high-confident unlabelled samples with lower loss values, thereby enhancing the contribution of such samples, (2) an inverse learning module that further improves the utilization of unlabelled data by learning what not to learn, and (3) a contrastive learning module that ensures learning from all the samples in both supervised and unsupervised settings. Comprehensive experiments on two popular 3D datasets demonstrate a performance improvement of up to 11.2% with 1% labelled data, surpassing the SOTA by a significant margin. Furthermore, AllMatch exhibits its efficiency in effectively leveraging all the unlabelled data, demonstrated by the fact that only 10% of labelled data reaches nearly the same performance as fully-supervised learning with all labelled data. The code of our work is available at: https://github.com/snehaputul/AllMatch.

Figures (6)

• Figure 1: Performance in comparison to unlabelled data usage.
• Figure 2: (left) Pseudo label utilization over the training epochs by FixMatch and ConFid. While ConFid does better utilization than FixMatch, none properly utilizes all unlabelled data. (right) Loss over training for high and low-confident pseudo-labels. The high-confident samples are selected for unsupervised learning, but low loss results in marginal contribution to learning.
• Figure 3: Impact of labelled set size.
• Figure 4: Sensitivity study on the augmentation strengths.
• Figure 5: Sensitivity study on the impact of different warm-up epochs.