Verified65% confidenceFactExact time
In 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.
3
Sources
65%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
Related Claims
UnverifiedIn convolutional neural networks (CNNs), patterns often correspond to visualizable feature maps, such as edge detectors or texture filters.73% similarUnverifiedExperiments reveal that the model's early layers are dominated by perceptual primitives, primarily processing low-level visual features such as color, shape, and texture.73% similarUnverifiedIn image classification, a CNN usually ends with a fully connected layer that outputs class probabilities; in object detection, the network predicts bounding box offsets and classes on the feature map; in image segmentation, an encoder-decoder structure is often used to upsample low-resolution semantic features back to the original image size to achieve pixel-level prediction.72% similarUnverified卷积神经网络（CNN）通过对大量标注植物图像的训练学会提取叶片纹理、颜色分布、边缘特征等视觉模式71% 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71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/336985API
curl https://kongchang.com/api/v1/knowledge/claims/336985MCP
get_claim(id=336985)