Object detectors are typically trained once and for all on a fixed set of classes. However, this closed-world assumption is unrealistic in practice, as new classes will inevitably emerge after the detector is deployed in the wild. In this work, we look at ways to extend a detector trained for a set of base classes so it can i) spot the presence of novel classes, and ii) automatically enrich its repertoire to be able to detect those newly discovered classes together with the base ones. We propose PANDAS, a method for novel class discovery and detection. It discovers clusters representing novel classes from unlabeled data, and represents old and new classes with prototypes. During inference, a distance-based classifier uses these prototypes to assign a label to each detected object instance. The simplicity of our method makes it widely applicable. We experimentally demonstrate the effectiveness of PANDAS on the VOC 2012 and COCO-to-LVIS benchmarks. It performs favorably against the state of the art for this task while being computationally more affordable.

翻译：目标检测器通常针对固定类别集合一次性训练完成。然而，这种封闭世界的假设在实践中并不现实，因为检测器部署到真实环境后，新类别不可避免地会出现。本研究探索如何扩展一个为基类训练的检测器，使其能够：i) 发现新类别的存在，ii) 自动丰富其检测能力，从而能够同时检测这些新发现的类别与基类。我们提出PANDAS方法，一种用于新类发现与检测的技术。该方法从无标签数据中发掘代表新类别的聚类，并使用原型表示新旧类别。推理阶段，基于距离的分类器利用这些原型为每个检测到的目标实例分配标签。我们方法的简洁性使其具有广泛适用性。实验证明，PANDAS在VOC 2012和COCO-to-LVIS基准上表现优异，在保持计算成本更低的同时，取得了优于该任务最先进方法的效果。