Automatic dataset generation for specific object detection
This addresses the scalability and information limitations in dataset generation for object detection tasks, though it appears incremental as it builds on existing synthesis and segmentation techniques.
The paper tackles the problem of inefficient and information-poor dataset creation for object detection by proposing a method to synthesize object-in-scene images that preserve detailed object features. The result shows that synthesized images blend objects well with backgrounds and enable state-of-the-art segmentation models to perform effectively.
In the past decade, object detection tasks are defined mostly by large public datasets. However, building object detection datasets is not scalable due to inefficient image collecting and labeling. Furthermore, most labels are still in the form of bounding boxes, which provide much less information than the real human visual system. In this paper, we present a method to synthesize object-in-scene images, which can preserve the objects' detailed features without bringing irrelevant information. In brief, given a set of images containing a target object, our algorithm first trains a model to find an approximate center of the object as an anchor, then makes an outline regression to estimate its boundary, and finally blends the object into a new scene. Our result shows that in the synthesized image, the boundaries of objects blend very well with the background. Experiments also show that SOTA segmentation models work well with our synthesized data.