Unverified50% confidenceFactExact time
Experiments reveal that the model's early layers are dominated by perceptual primitives, primarily processing low-level visual features such as color, shape, and texture.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedIn 2019, researchers at the University of Tübingen discovered that CNN models pre-trained on ImageNet rely more on texture features than shape features when recognizing objects, a phenomenon known as texture bias.73% similarUnverified经典CNN依赖局部感受野逐层提取从纹理到形状的层次化特征68% similarUnverifiedIn image recognition, shallow neurons respond to low-level features such as edges and textures, while deep neurons respond to overall concepts.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983114API
curl https://kongchang.com/api/v1/knowledge/claims/983114MCP
get_claim(id=983114)