Computational Baby Learning

Abstract

Intuitive observations show that a baby may inherently possess the capability of recognizing a new visual concept (e.g., chair, dog) by learning from only very few positive instances taught by parent(s) or others, and this recognition capability can be gradually further improved by exploring and/or interacting with the real instances in the physical world. Inspired by these observations, we propose a computational model for slightly-supervised object detection, based on prior knowledge modelling, exemplar learning and learning with video contexts. The prior knowledge is modeled with a pre-trained Convolutional Neural Network (CNN). When very few instances of a new concept are given, an initial concept detector is built by exemplar learning over the deep features from the pre-trained CNN. Simulating the baby's interaction with physical world, the well-designed tracking solution is then used to discover more diverse instances from the massive online unlabeled videos. Once a positive instance is detected/identified with high score in each video, more variable instances possibly from different view-angles and/or different distances are tracked and accumulated. Then the concept detector can be fine-tuned based on these new instances. This process can be repeated again and again till we obtain a very mature concept detector. Extensive experiments on Pascal VOC-07/10/12 object detection datasets well demonstrate the effectiveness of our framework. It can beat the state-of-the-art full-training based performances by learning from very few samples for each object category, along with about 20,000 unlabeled videos.

9 pages

References in corpus (4)

• Going Deeper with Convolutions
• CNN: Single-label to Multi-label
• Scalable Object Detection using Deep Neural Networks
• Associative embeddings for large-scale knowledge transfer with self-assessment

Cited by in corpus (12)

• STC: A Simple to Complex Framework for Weakly-supervised Semantic Segmentation
• Unsupervised Learning of Visual Representations using Videos
• BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation
• Self Paced Deep Learning for Weakly Supervised Object Detection
• Scale-aware Fast R-CNN for Pedestrian Detection
• Proposal-free Network for Instance-level Object Segmentation
• Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation
• Object Detection Networks on Convolutional Feature Maps
• Adaptive Object Detection Using Adjacency and Zoom Prediction
• Unsupervised Category Discovery via Looped Deep Pseudo-Task Optimization Using a Large Scale Radiology Image Database
• Unsupervised Joint Mining of Deep Features and Image Labels for Large-scale Radiology Image Categorization and Scene Recognition
• Object Recognition Based on Amounts of Unlabeled Data