Transfer Learning for Video Recognition with Scarce Training Data for Deep Convolutional Neural Network

TL;DR

This paper demonstrates that transfer learning from images to videos enables effective video recognition with limited training data, significantly reducing the need for large annotated video datasets.

Contribution

The authors propose a transfer learning approach from images to videos, enabling robust video recognition with only 4,000 videos and weakly labeled image data, reducing data requirements.

Findings

Transfer learning improves recognition accuracy on scarce data.

Only 4,000 videos needed for effective training.

Heuristic for meta-parameter selection developed.

Abstract

Unconstrained video recognition and Deep Convolution Network (DCN) are two active topics in computer vision recently. In this work, we apply DCNs as frame-based recognizers for video recognition. Our preliminary studies, however, show that video corpora with complete ground truth are usually not large and diverse enough to learn a robust model. The networks trained directly on the video data set suffer from significant overfitting and have poor recognition rate on the test set. The same lack-of-training-sample problem limits the usage of deep models on a wide range of computer vision problems where obtaining training data are difficult. To overcome the problem, we perform transfer learning from images to videos to utilize the knowledge in the weakly labeled image corpus for video recognition. The image corpus help to learn important visual patterns for natural images, while these patterns are ignored by models trained only on the video corpus. Therefore, the resultant networks have better generalizability and better recognition rate. We show that by means of transfer learning from image to video, we can learn a frame-based recognizer with only 4k videos. Because the image corpus is weakly labeled, the entire learning process requires only 4k annotated instances, which is far less than the million scale image data sets required by previous works. The same approach may be applied to other visual recognition tasks where only scarce training data is available, and it improves the applicability of DCNs in various computer vision problems. Our experiments also reveal the correlation between meta-parameters and the performance of DCNs, given the properties of the target problem and data. These results lead to a heuristic for meta-parameter selection for future researches, which does not rely on the time consuming meta-parameter search.