Learning Geo-Temporal Image Features
This work addresses the challenge of extracting mid-level features related to image capture context for applications in computer vision, though it is incremental as it builds on existing geo-temporal estimation methods.
The paper tackled the problem of learning geo-temporal image features by optimizing for location and time estimation tasks using a large dataset of outdoor images with known capture metadata, resulting in features that capture natural appearance changes and enable applications in time and location estimation.
We propose to implicitly learn to extract geo-temporal image features, which are mid-level features related to when and where an image was captured, by explicitly optimizing for a set of location and time estimation tasks. To train our method, we take advantage of a large image dataset, captured by outdoor webcams and cell phones. The only form of supervision we provide are the known capture time and location of each image. We find that our approach learns features that are related to natural appearance changes in outdoor scenes. Additionally, we demonstrate the application of these geo-temporal features to time and location estimation.