Instance Localization for Self-supervised Detection Pretraining
Prior research on self-supervised learning has led to considerable progress on image classification, but often with degraded transfer performance on object detection. The objective of this paper is to advance self-supervised pretrained models specifically for object detection. Based on the inherent difference between classification and detection, we propose a new self-supervised pretext task, called instance localization. Image instances are pasted at various locations and scales onto background images. The pretext task is to predict the instance category given the composited images as well as the foreground bounding boxes. We show that integration of bounding boxes into pretraining promotes better task alignment and architecture alignment for transfer learning. In addition, we propose an augmentation method on the bounding boxes to further enhance the feature alignment. As a result, our model becomes weaker at Imagenet semantic classification but stronger at image patch localization, with an overall stronger pretrained model for object detection. Experimental results demonstrate that our approach yields state-of-the-art transfer learning results for object detection on PASCAL VOC and MSCOCO.
Code (1)
Tasks
ClassificationGeneral Classificationimage-classificationImage ClassificationObjectobject-detectionObject DetectionSelf-Supervised LearningTransfer LearningSimilar Papers 제목 키워드 기반
DETReg: Unsupervised Pretraining with Region Priors for Object Detection
Recent self-supervised pretraining methods for object detection largely focus on pretraining the backbone of the object detector, neglecting key parts of detection architecture. Instead, we introduce DETReg, a new self-s…
Few-Shot LearningFew-Shot Object DetectionObjectobject-detection+5Spatially Consistent Representation Learning
Self-supervised learning has been widely used to obtain transferrable representations from unlabeled images. Especially, recent contrastive learning methods have shown impressive performances on downstream image classifi…
Contrastive Learningimage-classificationImage ClassificationInstance Segmentation+5Skeleton-based Zero-Shot Spatio-Temporal Action Localization via Weakly-Supervised Pretraining
We propose a novel pretraining strategy for skeleton-based zero-shot spatio-temporal action localization to estimate unseen actions for person instances while overcoming high annotation costs for training via new target …
Spatio-Temporal Action LocalizationContrastive LearningTowards Accurate Localization by Instance Search
Visual object localization is the key step in a series of object detection tasks. In the literature, high localization accuracy is achieved with the mainstream strongly supervised frameworks. However, such methods requir…
Few-Shot Object DetectionInstance SearchObjectobject-detection+2Benchmarking Self-Supervised Contrastive Learning Methods for Image-Based Plant Phenotyping
The rise of self-supervised learning (SSL) methods in recent years presents an opportunity to leverage unlabeled and domain-specific datasets generated by image-based plant phenotyping platforms to accelerate plant breed…
BenchmarkingContrastive LearningHead DetectionPlant Phenotyping+1