|国家预印本平台
首页|Learning More by Seeing Less: Line Drawing Pretraining for Efficient, Transferable, and Human-Aligned Vision

Learning More by Seeing Less: Line Drawing Pretraining for Efficient, Transferable, and Human-Aligned Vision

Learning More by Seeing Less: Line Drawing Pretraining for Efficient, Transferable, and Human-Aligned Vision

来源:Arxiv_logoArxiv
英文摘要

Despite remarkable progress in computer vision, modern recognition systems remain limited by their dependence on rich, redundant visual inputs. In contrast, humans can effortlessly understand sparse, minimal representations like line drawings - suggesting that structure, rather than appearance, underlies efficient visual understanding. In this work, we propose using line drawings as a structure-first pretraining modality to induce more compact and generalizable visual representations. We show that models pretrained on line drawings develop stronger shape bias, more focused attention, and greater data efficiency across classification, detection, and segmentation tasks. Notably, these models also exhibit lower intrinsic dimensionality, requiring significantly fewer principal components to capture representational variance - echoing the similar observation in low dimensional efficient representation in the brain. Beyond performance improvements, line drawing pretraining produces more compressible representations, enabling better distillation into lightweight student models. Students distilled from line-pretrained teachers consistently outperform those trained from color-supervised teachers, highlighting the benefits of structurally compact knowledge. Finally, we demonstrate that the pretraining with line-drawing can also be extended to unsupervised setting via our proposed method "learning to draw". Together, our results support the view that structure-first visual learning fosters efficiency, generalization, and human-aligned inductive biases - offering a simple yet powerful strategy for building more robust and adaptable vision systems.

Tianqin Li、George Liu、Tai Sing Lee

信息科学、信息技术自然科学研究方法

Tianqin Li,George Liu,Tai Sing Lee.Learning More by Seeing Less: Line Drawing Pretraining for Efficient, Transferable, and Human-Aligned Vision[EB/OL].(2025-08-08)[2025-08-24].https://arxiv.org/abs/2508.06696.点此复制

评论