We proposed an attentional network for visual object detection. It sequentially generated glimpse regions of various sizes and aspect ratios, extracted features from these regions, and made a final decision based on the information it had acquired...
Over the years, computer vision researchers have spent an immense amount of effort on designing image features for the visual object recognition task. We propose to incorporate this valuable experience to guide the task of training deep neural net...
Credentials & highlights
- Project Views