Abstract: Traditional deep-learning-based object detection networks often resize images during the data preprocessing stage to achieve a uniform size and scale in the feature map. Resizing is done to ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results