ArticleVision research2020
Local features and global shape information in object classification by deep convolutional neural networks.
Article in Vision research, 2020. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 26 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
26 citing papers in PubMed, 88 citations in OpenAlex.
- Systematic image perturbations reveal persistent gaps between human and machine vision.iScience · 2026Article
- Self-supervised learning yields representational signatures of category-selective cortex.Communications biology · 2026Article
- Shape-similarity gain: A selection profile for simple objects.Attention, perception & psychophysics · 2026Article
- Hierarchical abstraction drives human-like 3-D shape processing in deep learning models.PLoS computational biology · 2026Article
- Three-dimensional shape cues affect human and artificial recognition systems differently.PloS one · 2026Article
- Vision Transformer attention alignment with human visual perception in aesthetic object evaluation.PloS one · 2026Article
- Mitigating data bias and ensuring reliable evaluation of AI models with shortcut hull learning.Nature communications · 2025Article
- Comparing Activation Typicality and Sparsity in a Deep CNN to Predict Facial Beauty.Computational brain & behavior · 2025Article
- Configural processing as an optimized strategy for robust object recognition in neural networks.Communications biology · 2025Article
- Memorability of line drawings of scenes: the role of contour properties.Memory & cognition · 2025Article
- IDNet: An inception-like deformable non-local network for projection compensation over non-flat textured surfaces.PloS one · 2025Article
- Deep convolutional neural networks are sensitive to face configuration.Journal of vision · 2024Article
- Teaching deep networks to see shape: Lessons from a simplified visual world.PLoS computational biology · 2024Article
- Population encoding of stimulus features along the visual hierarchy.Proceedings of the National Academy of Sciences of the United States of America · 2024Article
- Object geometry serves humans' intuitive physics of stability.Scientific reports · 2024Article
- Sparsity in an artificial neural network predicts beauty: Towards a model of processing-based aesthetics.PLoS computational biology · 2023Article
- 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
- Population encoding of stimulus features along the visual hierarchy.bioRxiv : the preprint server for biology · 2023Article
- Deep learning models fail to capture the configural nature of human shape perception.iScience · 2022Article
- General object-based features account for letter perception.PLoS computational biology · 2022Article
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 at 1 institution in 1 country.
Funding
Abstract
Deep convolutional neural networks (DCNNs) show impressive similarities to the human visual system. Recent research, however, suggests that DCNNs have limitations in recognizing objects by their shape. We tested the hypothesis that DCNNs are sensitive to an object's local contour features but have no access to global shape information that predominates human object recognition. We employed transfer learning to assess local and global shape processing in trained networks. In Experiment 1, we used restricted and unrestricted transfer learning to retrain AlexNet, VGG-19, and ResNet-50 to classify circles and squares. We then probed these networks with stimuli with conflicting global shape and local contour information. We presented networks with overall square shapes comprised of curved elements and circles comprised of corner elements. Networks classified the test stimuli by local contour features rather than global shapes. In Experiment 2, we changed the training data to include circles and squares comprised of different elements so that the local contour features of the object were uninformative. This considerably increased the network's tendency to produce global shape responses, but deeper analyses in Experiment 3 revealed the network still showed no sensitivity to the spatial configuration of local elements. These findings demonstrate that DCNNs' performance is an inversion of human performance with respect to global and local shape processing. Whereas abstract relations of elements predominate in human perception of shape, DCNNs appear to extract only local contour fragments, with no representation of how they spatially relate to each other to form global shapes.
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.