The abstract is read at the publisher; the summary is JClub's.
Humans identify common objects with the same accuracy using local visual features as when using shape, in seven experiments with adults (N = 300) and 5-year-old children (N = 100).
- Why it matters: The work addresses whether shape is truly essential for human object perception, given that leading DNN vision models often depend more on texture-like local cues than on shape.
- What they did: Across seven experiments, participants viewed objects manipulated to emphasize either shape or local features, testing recognition and learning in adults (N = 300) and 5-year-old children (N = 100).
- The result: Shape is crucial for learning novel objects, enabling even one-shot object categorization, implying local features may suffice for everyday recognition but not for rapid learning of new categories.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.