Position, Padding and Predictions: A Deeper Look at Position Information in CNNs

This stduy explores how to utilize position information in CNN models for down-stream tasks.

Shape or texture: Understanding Discriminative Features in CNNs

This study quantitatively measures the shape and texture information encoded by a CNN model.

Global Pooling, More than Meets the Eye: Position Information is Encoded Channel-Wise in CNNs

This study explores how the position information is encoded in a CNN model.

Revisiting Saliency Metrics: Farthest-Neighbor Area Under Curve

This study proposes a new metric for saliency detection, which can better overcome the center bias problem.

How Much Position Information Do Convolutional Neural Networks Encode?

This is the first study to explore position information implicitly learned by a CNN model.

Artificial intelligence device for harvesting data from unlabeled sources and control method thereof

Artificial intelligence device for evaluation and control method thereof

Device for training and managing a visual scene graph model and control method thereof

Device for managing a visual saliency model and control method thereof