PCL: Proposal Cluster Learning for Weakly Supervised Object Detection

By Peng Tang, Xinggang Wang, Song Bai, Wei Shen, Xiang Bai, Alan L. Yuille. 2018

405 citations recorded by the source corpus.

Weakly Supervised Object Detection (WSOD), using only image-level annotations to train object detectors, is of growing importance in object recognition. In this paper, we propose a novel deep network for WSOD. Unlike previous networks that transfer the object detection problem to an image classification problem using Multiple Instance Learning (MIL), our strategy generates proposal clusters to learn refined instance classifiers by an iterative process. The proposals in the same cluster are spatially adjacent and associated with the same object. This prevents the network from concentrating too much on parts of objects instead of whole objects. We first show that instances can be assigned object or background labels directly based on proposal clusters for instance classifier refinement, and then show that treating each cluster as a small new bag yields fewer ambiguities than the directly assigning label method. The iterative instance classifier refinement is implemented online using multiple streams in convolutional neural networks, where the first is an MIL network and the others are for instance classifier refinement supervised by the preceding one. Experiments are conducted on the PASCAL VOC, ImageNet detection, and MS-COCO benchmarks for WSOD. Results show that our method outperforms the previous state of the art significantly.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Evaluation & Benchmarks

Citations in the corpus

Learning Open-World Object Proposals without Learning to Classify (2021, background): “They have become the integral pre-processing steps for many computer vision systems, including object detection [33,41,42,45,59,74], segmentation [3,8,13], object discovery [12, 16, 47], weakly supervised object detection [5, 21, 53], visual tracking [35, 63], content-aware retargeting [51], etc.”
Open Vocabulary Object Detection with Pseudo Bounding-Box Labels (2021, background): “Weakly supervised object detectors seek to relieve such heavy human annotation burden by using image-level labels such as image-level object categories [2,29], captions [32] and object counts [7] for training.”
A Survey of Deep Learning-Based Object Detection (2019, methodology): “[229] utilize an iterative procedure to generate proposal clusters and learn refined instance classifiers, which makes the network concentrate on the whole object rather than part of it.”
From Discriminant to Complete: Reinforcement Searching-Agent Learning for Weakly Supervised Object Detection (2020, result): “When compared to Tang’s (L) [54], our approach obtains slightly worse performance on the trainval data sets but better performance on the test data sets.”
Detecting Twenty-thousand Classes using Image-level Supervision (2022, methodology): “DLWL [39] combines self-training and clustering-based WSOD [53], and again assigns image labels to max-scored proposals.”
What leads to generalization of object proposals? (2020, background): “Object detection systems have shown considerable improvements for fully supervised settings [27,18,20,26,3], as well as weakly supervised settings [7,1,32] that only use image-level labels.”
Object Detection in 20 Years: A Survey (2019, neutral): “Multi-instance learning is a group of supervised learning algorithms that has seen widespread application in WSOD [202], [203], [204], [205], [206], [207], [208], [209].”
Product1M: Towards Weakly Supervised Instance-Level Product Retrieval via Cross-Modal Pretraining (2021, background): “PCL [39] iteratively generates proposal clusters to facilitate the learning of instance classifiers.”
C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object Detection (2019, methodology): “Instead of directly labeling the positive proposals to class c , we treat these positive proposals as a bag, and use a averaged MIL pooling method [30].”
Min-Entropy Latent Model for Weakly Supervised Object Detection (2018, methodology): “The iterative generation of the proposal clusters [43] with OICR prevented the network from concentrating on parts of objects.”
Weakly Supervised Object Localization and Detection: A Survey (2021, neutral): “Tang-TPAMI-2018 [145] Fast RCNN* (model ensemble) CNN None ImageNet(tag label) Fast RCNN* (model ensemble) MIL+oicr+multi-scale+proposal cluster learning General objects”
Omni-DETR: Omni-Supervised Object Detection with Transformers (2022, positive): “This establishes a much stronger baseline than any previous weakly supervised object detection (WSOD) and weakly semi-supervised object detection (WSSOD) work [13, 23, 32, 44].”
Toward Self-Supervised Object Detection in Unlabeled Videos (2019, methodology): “As our baseline, we opted for the PCL weakly-supervised object detection method [22] currently achieving the highest mAP of 19.”
Learning to Detect and Retrieve Objects From Unlabeled Videos (2019, methodology): “As our baseline, we opted for the Proposal Cluster Learning (PCL) WSOD method [15] currently achieving the highest mAP of 19.6% for WSOD on ImageNet-Det.”
Comprehensive Attention Self-Distillation for Weakly-Supervised Object Detection (2020, methodology): “To have a fair comparison with other methods, following [12, 7], multi-level scaling and horizontal flipping data augmentation are conducted in training.”