Image amodal completion: A survey

Existing computer vision systems can compete with humans in understanding the visible parts of objects, but still fall far short when it comes to depicting invisible partially occluded objects. Image amodal completion aims equip computers human-like functions understand an intact object despite being occluded. The main purpose this survey is provide intuitive research hotspots, key technologies and future trends field image completion. Firstly, we present a comprehensive review latest literature emerging field, exploring three tasks completion, including shape appearance order perception. Then examine popular datasets related along their common data collection methods evaluation metrics. Finally, discuss real-world applications directions for facilitating reader’s challenges existing upcoming trends.