ArticleJournal of vision2022
From photos to sketches - how humans and deep neural networks process objects across different levels of visual abstraction.
Article in Journal of vision, 2022. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 16 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
16 citing papers in PubMed.
- Optimizing Deep Learning for Renal Mass Characterization in Challenging Cases: A Comparative Study of Spatial-Input Strategies and a Validated ROI-Only DLR Nomogram.Annals of surgical oncology · 2026Article
- Can neural networks model the human perception of geometric shapes?PLoS computational biology · 2026Article
- Tracing aesthetic experience from perception and conception to appraisal using deep convolutional neural networks.iScience · 2026Article
- Drawings of THINGS: A large-scale drawing dataset of 1854 object concepts.Behavior research methods · 2026Article
- Dimensions underlying the representational alignment of deep neural networks with humans.Nature machine intelligence · 2025Article
- Using drawings and deep neural networks to characterize the building blocks of human visual similarity.Memory & cognition · 2025Article
- Emergence of Emotion Selectivity in Deep Neural Networks Trained to Recognize Visual Objects.bioRxiv : the preprint server for biology · 2024Article
- Emergence of Emotion Selectivity in Deep Neural Networks Trained to Recognize Visual Objects.PLoS computational biology · 2024Article
- Model metamers reveal divergent invariances between biological and artificial neural networks.Nature neuroscience · 2023Article
- Evaluation of the Hierarchical Correspondence between the Human Brain and Artificial Neural Networks: A Review.Biology · 2023Review
- A novel feature-scrambling approach reveals the capacity of convolutional neural networks to learn spatial relations.Neural networks : the official journal of the International Neural Network Society · 2023Article
- Drawing as a versatile cognitive tool.Nature reviews psychology · 2023Article
- The neuroconnectionist research programme.Nature reviews. Neuroscience · 2023Review
- The Spatiotemporal Neural Dynamics of Object Recognition for Natural Images and Line Drawings.The Journal of neuroscience : the official journal of the Society for Neuroscience · 2023Article
- Deep learning for studying drawing behavior: A review.Frontiers in psychology · 2023Review
- Investigating the temporal dynamics and modeling of mid-level feature representations in humans.Imaging neuroscience (Cambridge, Mass.)Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
4 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Line drawings convey meaning with just a few strokes. Despite strong simplifications, humans can recognize objects depicted in such abstracted images without effort. To what degree do deep convolutional neural networks (CNNs) mirror this human ability to generalize to abstracted object images? While CNNs trained on natural images have been shown to exhibit poor classification performance on drawings, other work has demonstrated highly similar latent representations in the networks for abstracted and natural images. Here, we address these seemingly conflicting findings by analyzing the activation patterns of a CNN trained on natural images across a set of photographs, drawings, and sketches of the same objects and comparing them to human behavior. We find a highly similar representational structure across levels of visual abstraction in early and intermediate layers of the network. This similarity, however, does not translate to later stages in the network, resulting in low classification performance for drawings and sketches. We identified that texture bias in CNNs contributes to the dissimilar representational structure in late layers and the poor performance on drawings. Finally, by fine-tuning late network layers with object drawings, we show that performance can be largely restored, demonstrating the general utility of features learned on natural images in early and intermediate layers for the recognition of drawings. In conclusion, generalization to abstracted images, such as drawings, seems to be an emergent property of CNNs trained on natural images, which is, however, suppressed by domain-related biases that arise during later processing stages in the network.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.