Learning Visual Features from Large Weakly Supervised Data
arXiv:1511.02251
Abstract
Convolutional networks trained on large supervised dataset produce visual features which form the basis for the state-of-the-art in many computer-vision problems. Further improvements of these visual features will likely require even larger manually labeled data sets, which severely limits the pace at which progress can be made. In this paper, we explore the potential of leveraging massive, weakly-labeled image collections for learning good visual features. We train convolutional networks on a dataset of 100 million Flickr photos and captions, and show that these networks produce features that perform well in a range of vision problems. We also show that the networks appropriately capture word similarity, and learn correspondences between different languages.
References in corpus (11)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Efficient Estimation of Word Representations in Vector Space
- Going Deeper with Convolutions
- VQA: Visual Question Answering
- Deep Fragment Embeddings for Bidirectional Image Sentence Mapping
- Learning to Segment Object Candidates
- CNN Features off-the-shelf: an Astounding Baseline for Recognition
- Rich feature hierarchies for accurate object detection and semantic segmentation
- Webly Supervised Learning of Convolutional Networks
- Large-Scale Deep Learning on the YFCC100M Dataset
- ConceptLearner: Discovering Visual Concepts from Weakly Labeled Image Collections
Cited by in corpus (5)
- Revisiting Unreasonable Effectiveness of Data in Deep Learning Era
- Efficient softmax approximation for GPUs
- Leveraging Weakly Annotated Data for Fashion Image Retrieval and Label Prediction
- Deep Learning from Noisy Image Labels with Quality Embedding
- Conditional Denoising of Remote Sensing Imagery Using Cycle-Consistent Deep Generative Models