Abstract
We propose Cut-and-LEaRn (CutLER), a simple approach for training unsupervised object detection and segmentation models. We leverage the property of self-supervised models to 'discover' objects without supervision and amplify it to train a state-of-the-art localization model without any human labels. CutLER first uses our proposed MaskCut approach to generate coarse masks for multiple objects in an image and then learns a detector on these masks using our robust loss function. We further improve the performance by self-training the model on its predictions. Compared to prior work, CutLER is simpler, compatible with different detection architectures, and detects multiple objects. CutLER is also a zero-shot unsupervised detector and improves detection performance AP50 by over 2.7 times on 11 benchmarks across domains like video frames, paintings, sketches, etc. With finetuning, CutLER serves as a low-shot detector surpassing MoCo-v2 by 7.3% APbox and 6.6% APmask on COCO when training with 5% labels.
Abstract (translated)
我们提出了 Cut-and-LEaRn (Cutler) 一种简单的方法,用于训练无监督物体检测和分割模型。我们利用自监督模型的特性,在没有监督的情况下“发现”物体,并放大它来训练没有人类标签的最先进的定位模型。 Cutler 先使用我们提出的MaskCut方法生成图像中的粗面具,然后使用我们的稳健损失函数学习对这些面具的探测器。我们通过自训练模型来进一步改善性能。与以前的工作相比, Cutler 更简单,与不同的检测架构兼容,能够检测多个物体。 Cutler 也是零样本无监督探测器,在包括视频帧、画作、 Sketch 等不同领域的11个基准上,提高了检测性能AP50超过2.7倍。通过微调, Cutler 成为低样本探测器,在训练时通过5%的标注数据超越MoCo-v2,在COCO中超过COCO-v2的APbox和APmask。
URL
https://arxiv.org/abs/2301.11320