Fine-to-coarse Knowledge Transfer For Low-Res Image Classification
This addresses the challenge of fine-grained classification in low-resolution images, such as in surveillance or satellite imagery, with incremental improvements in knowledge transfer methods.
The paper tackles the problem of distinguishing fine-grained object categories in low-resolution images by proposing a deep learning approach that transfers knowledge from high-resolution training data to low-resolution test scenarios, achieving effective transfer on benchmark datasets for car models and bird species.
We address the difficult problem of distinguishing fine-grained object categories in low resolution images. Wepropose a simple an effective deep learning approach that transfers fine-grained knowledge gained from high resolution training data to the coarse low-resolution test scenario. Such fine-to-coarse knowledge transfer has many real world applications, such as identifying objects in surveillance photos or satellite images where the image resolution at the test time is very low but plenty of high resolution photos of similar objects are available. Our extensive experiments on two standard benchmark datasets containing fine-grained car models and bird species demonstrate that our approach can effectively transfer fine-detail knowledge to coarse-detail imagery.