Improving Object Detection with Selective Self-supervised Self-training

  title={Improving Object Detection with Selective Self-supervised Self-training},
  author={Yandong Li and Di Huang and Danfeng Qin and Liqiang Wang and Boqing Gong},
We study how to leverage Web images to augment human-curated object detection datasets. Our approach is two-pronged. On the one hand, we retrieve Web images by image-to-image search, which incurs less domain shift from the curated data than other search methods. The Web images are diverse, supplying a wide variety of object poses, appearances, their interactions with the context, etc. On the other hand, we propose a novel learning method motivated by two parallel lines of work that explore… Expand
Instant-Teaching: An End-to-End Semi-Supervised Object Detection Framework
InstantTeaching is proposed, a completely end-to-end and effective SSOD framework, which uses instant pseudo labeling with extended weak-strong data augmentations for teaching during each training iteration, and further proposes a co-rectify scheme based on InstantTeaching, denoted as Instant-Teaching. Expand
Data-efficient Weakly-supervised Learning for On-line Object Detection under Domain Shift in Robotics
This work compares several techniques for weakly-supervised learning in detection pipelines to reduce model (re)training costs without compromising accuracy, and shows that diversity sampling for constructing active learning queries and strong positives selection for selfsupervisedLearning enable significant annotation savings and improve domain shift adaptation. Expand
End-to-End Semi-Supervised Object Detection with Soft Teacher
This paper presents an end-to-end semi-supervised object detection approach, in contrast to previous more complex multi-stage methods, and proposes two simple yet effective techniques within this framework: a soft teacher mechanism where the classification loss of each unlabeled bounding box is weighed by the classification score produced by the teacher network. Expand
Self-supervised Learning of 3D Object Understanding by Data Association and Landmark Estimation for Image Sequence
A self-supervised learning method to exploit multiple observations of the object in the image sequence in order to surpass the self-performance and network fine-tuning is conducted including the dataset obtained by self-annotation, thereby exceeding the performance boundary of the network itself. Expand
Learning to Better Segment Objects from Unseen Classes with Unlabeled Videos
This paper introduces a Bayesian method that is specifically designed to automatically create a high-quality training set which significantly boosts the performance of segmenting objects of unseen classes and could open the door for open-world instance segmentation using abundant Internet videos. Expand
Multi-Object Tracking with Hallucinated and Unlabeled Videos
In this paper, we explore learning end-to-end deep neural trackers without tracking annotations. This is important as large-scale training data is essential for training deep neural trackers whileExpand
Weakly-Supervised Object Detection Learning through Human-Robot Interaction
This paper presents a pipeline for efficiently training an object detection system on a humanoid robot by exploiting a teacher-learner pipeline, weakly supervised learning techniques to reduce the human labeling effort, and an on-line learning approach for fast model re-training. Expand
A Simple and Effective Use of Object-Centric Images for Long-Tailed Object Detection
The approach turns an object-centric image into a useful training example for object detection in scene-centric images by mitigating the domain gap between the two image sources in both the input and label space and employs a multi-stage procedure to train the object detector. Expand
Contrastive Learning with Temporal Correlated Medical Images: A Case Study using Lung Segmentation in Chest X-Rays
Contrastive learning has been proved to be a promising technique for image-level representation learning from unlabeled data. Many existing works have demonstrated improved results by applyingExpand


Zero-Annotation Object Detection with Web Knowledge Transfer
This work proposes an object detection method that does not require any form of human annotation on target tasks, by exploiting freely available web images, and introduces a multi-instance multi-label domain adaption learning framework with two key innovations. Expand
Towards Human-Machine Cooperation: Self-Supervised Sample Mining for Object Detection
This paper presents a principled Self-supervised Sample Mining (SSM) process accounting for the real challenges in object detection, and proposes a new AL framework for gradually incorporating unlabeled or partially labeled data into the model learning while minimizing the annotating effort of users. Expand
Towards Precise End-to-End Weakly Supervised Object Detection Network
This paper designs a single network with both multiple instance learning and bounding-box regression branches that share the same backbone and adds a guided attention module using classification loss to the backbone for effectively extracting the implicit location information in the features. Expand
Multi-Task Self-Supervised Object Detection via Recycling of Bounding Box Annotations
A novel object detection approach that takes advantage of both multi-task learning (MTL) and self-supervised learning (SSL) to improve the accuracy of object detection and empirically validate that this approach effectively improves detection performance on various architectures and datasets. Expand
Weakly- and Semi-Supervised Object Detection with Expectation-Maximization Algorithm
This paper develops an Expectation-Maximization (EM) based object detection method using deep convolutional neural networks (CNNs) that can almost match the performace of the fully supervised Fast RCNN. Expand
Weakly Supervised Object Detection With Segmentation Collaboration
This work proposes a novel end-to-end weakly supervised detection approach, where a newly introduced generative adversarial segmentation module interacts with the conventional detection module in a collaborative loop, forming a more comprehensive solution. Expand
Large Scale Semi-Supervised Object Detection Using Visual and Semantic Knowledge Transfer
Strong evidence is found that visual similarity and semantic relatedness are complementary for the task, and when combined notably improve detection, achieving state-of-the-art detection performance in a semi-supervised setting. Expand
Detecting 11K Classes: Large Scale Object Detection Without Fine-Grained Bounding Boxes
  • Hao Yang, Hao-Yu Wu, Hao Chen
  • Computer Science
  • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
  • 2019
This paper proposes a semi-supervised large scale fine-grained detection method, which only needs bounding box annotations of a smaller number of coarse- grained classes and image-level labels of large scalefine-grains classes, and can detect all classes at nearly fully-super supervised accuracy. Expand
Activity Driven Weakly Supervised Object Detection
This work shows that the action depicted in the image/video can provide strong cues about the location of the associated object and learns a spatial prior for the object dependent on the action, and incorporates this prior to simultaneously train a joint object detection and action classification model. Expand
NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection
A novel NOise Tolerant Ensemble RCNN (NOTE-RCNN) object detector is proposed, in which a few seed box level annotations and a large scale of image level annotations are used to train the detector. Expand