Geo-Spatiotemporal Features and Shape-Based Prior Knowledge for Fine-grained Imbalanced Data Classification
This work addresses a domain-specific problem for wildlife experts, but appears incremental as it combines existing innovations.
The paper tackled fine-grained classification in wildlife by combining geo-spatiotemporal features and shape-based priors to address small inter-class and large intra-class variations, resulting in improved performance with unspecified numbers.
Fine-grained classification aims at distinguishing between items with similar global perception and patterns, but that differ by minute details. Our primary challenges come from both small inter-class variations and large intra-class variations. In this article, we propose to combine several innovations to improve fine-grained classification within the use-case of wildlife, which is of practical interest for experts. We utilize geo-spatiotemporal data to enrich the picture information and further improve the performance. We also investigate state-of-the-art methods for handling the imbalanced data issue.