Few-Example Object Detection with Model Communication

Abstract

In this paper, we study object detection using a large pool of unlabeled images and only a few labeled images per category, named "few-example object detection". The key challenge consists in generating trustworthy training samples as many as possible from the pool. Using few training examples as seeds, our method iterates between model training and high-confidence sample selection. In training, easy samples are generated first and, then the poorly initialized model undergoes improvement. As the model becomes more discriminative, challenging but reliable samples are selected. After that, another round of model improvement takes place. To further improve the precision and recall of the generated training samples, we embed multiple detection models in our framework, which has proven to outperform the single model baseline and the model ensemble method. Experiments on PASCAL VOC'07, MS COCO'14, and ILSVRC'13 indicate that by using as few as three or four samples selected for each category, our method produces very competitive results when compared to the state-of-the-art weakly-supervised approaches using a large number of image-level labels.

Accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) 2018

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Active Self-Paced Learning for Cost-Effective and Progressive Face Identification
• Zero-Shot Detection
• Weakly-supervised Discovery of Visual Pattern Configurations
• Discriminatively Trained And-Or Graph Models for Object Shape Detection
• Bridging Saliency Detection to Weakly Supervised Object Detection Based on Self-paced Curriculum Learning
• Weakly Supervised Object Localization Using Size Estimates

Cited by in corpus (32)

• Delta-encoder: an effective sample synthesis method for few-shot object recognition
• Deep Learning for Generic Object Detection: A Survey
• HA-CCN: Hierarchical Attention-based Crowd Counting Network
• Few-shot Object Detection on Remote Sensing Images
• Adversarial Complementary Learning for Weakly Supervised Object Localization
• One-Shot Instance Segmentation
• Small Sample Learning in Big Data Era
• OSLNet: Deep Small-Sample Classification with an Orthogonal Softmax Layer
• Learning Non-Metric Visual Similarity for Image Retrieval
• Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector
• Recent Advances in Deep Learning for Object Detection
• Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection
• Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
• Attention-based Dropout Layer for Weakly Supervised Object Localization
• DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detection
• Few-shot Object Detection via Feature Reweighting
• A Comparative Review of Recent Few-Shot Object Detection Algorithms
• Hallucination Improves Few-Shot Object Detection
• Revisiting Few-shot Activity Detection with Class Similarity Control
• Self-produced Guidance for Weakly-supervised Object Localization
• MM-FSOD: Meta and metric integrated few-shot object detection
• Teacher Supervises Students How to Learn From Partially Labeled Images for Facial Landmark Detection
• Comparison-Based Convolutional Neural Networks for Cervical Cell/Clumps Detection in the Limited Data Scenario
• Universal Lesion Detection by Learning from Multiple Heterogeneously Labeled Datasets
• Few-shot Adaptive Faster R-CNN
• A Concise Review of Recent Few-shot Meta-learning Methods
• Semi-supervised Anatomical Landmark Detection via Shape-regulated Self-training
• Towards Human-Machine Cooperation: Self-supervised Sample Mining for Object Detection
• Training Object Detectors from Few Weakly-Labeled and Many Unlabeled Images
• Localizing the Common Action Among a Few Videos
• Leveraging Pre-Trained 3D Object Detection Models For Fast Ground Truth Generation
• Bootstrap Your Object Detector via Mixed Training