TACoS: Weakly Supervised Learning of Two-Dimensional Materials from Scribble Annotations to Precise Segmentation

2026-07-08 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Physical Sciences & Chemistry · Depth: Expert, quick

Summary

TACoS is a specialized scribble segmentation framework designed for precise pixel-level localization of two-dimensional (2D) material flakes, addressing the high cost and time consumption of dense annotations in traditional fully supervised methods. This framework integrates semi-supervised consistency learning with structured tree energy constraints, featuring an unlabeled weak-strong distribution alignment module using cosine consistency and a tree energy regularization module that employs minimum spanning trees for pixel affinity and structure-aware pseudo labels. Additionally, TACoS introduces asymmetric regional contrast learning, which fuses high-confidence predictions with scribbles to form augmented labels and constructs category prototypes. This strategy prioritizes contrastive constraints on challenging pixels, enhancing intra-class cohesion and inter-class separation. Experiments on graphene and MoS2 datasets show TACoS achieves over 96% of fully supervised performance with less than 0.6% annotated data, demonstrating superior structural coherence and boundary stability.

Key takeaway

For AI Scientists and Research Scientists developing segmentation models for 2D materials, TACoS offers a highly efficient alternative to fully supervised methods. You can achieve over 96% of fully supervised performance with less than 0.6% of the annotation effort. Consider integrating similar semi-supervised consistency learning and asymmetric regional contrast techniques to significantly reduce data labeling costs and accelerate high-throughput screening.

Key insights

TACoS enables precise 2D material segmentation using minimal scribble annotations via a novel semi-supervised framework.

Principles

• Integrate consistency learning with structural constraints.
• Prioritize contrastive learning on challenging boundary pixels.

Method

TACoS combines weak-strong distribution alignment with tree energy regularization for pseudo-label generation, then uses asymmetric regional contrast learning to refine representations.

In practice

• Automate high-throughput screening of 2D material flakes.
• Reduce annotation burden for material science segmentation.

Topics

• Weakly Supervised Learning
• 2D Materials
• Image Segmentation
• Semi-supervised Learning
• Contrastive Learning
• High-throughput Screening

Best for: AI Scientist, Research Scientist

Related on AIssential

• Learning to Label: A Reinforced Self-Evolving Framework for Semi-supervised Referring Expression Segmentation — A reinforced self-evolving framework improves semi-supervised referring expression segmentation by learning to construct reliable pseudo-labels.
• Complex Layout Classification in the Wild: A Low-Resource Approach with Layout-Preserving Augmentations — Domain-aware augmentations significantly improve low-resource complex layout classification by focusing on global geometric arrangements.
• SegMix:Shuffle-based Feedback Learning for Semantic Segmentation of Pathology Images — Shuffle-based feedback learning improves weakly supervised semantic segmentation in pathology by adapting to multi-scale image characteristics.
• Accuracy Improvement of Semi-Supervised Segmentation Using Supervised ClassMix and Sup-Unsup Feature Discriminator — Improved semi-supervised segmentation uses supervised ClassMix and a feature discriminator to enhance accuracy and mitigate pseudo-label inaccuracies.
• A Review of Pseudo-Labeling for Computer Vision — Pseudo-labeling extends beyond semi-supervised learning to enhance self-supervised and unsupervised methods.
• Structure-Semantic Co-optimized Latent Diffusion Model for Fast Visual Anagram Synthesis — S2CO-Anagram improves visual anagram synthesis via co-optimized latent diffusion for quality and speed.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.