Deep reinforcement learning-based captioning with embedding reward

• US 10,726,059 B1
• Filed: 06/21/2019
• Issued: 07/28/2020
• Est. Priority Date: 11/10/2016
• Status: Active Grant

• Current Assignee: Snap, Inc.
• Original Assignee: Snap, Inc.
• Inventors: Ren, Zhou, Wang, Xiaoyu, Zhang, Ning, Lv, Xutao, Li, Jia
• Primary Examiner(s): Liew, Alex Kok S
• Application Number:
  US16/448,900
• Time in Patent Office: 403 Days
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06N 3/006 based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06V 10/811 the classifiers operating on different input data, e.g. multi-modal recognition
• CPC Class Codes: G06V 10/82 using neural networks
• CPC Class Codes: G06V 10/94 Hardware or software architectures specially adapted for image or video understanding

Abstract

An image captioning system and method is provided for generating a caption for an image. The image captioning system utilizes a policy network and a value network to generate the caption. The policy network serves as a local guidance and the value network serves as a global and lookahead guidance.

Claims

1. A method comprising:

extracting, by an image captioning system, an image feature from an image; and

analyzing the image feature to generate a caption describing the image feature by performing the following operations until the caption describing the image feature is complete;

analyzing, by a policy network of the image captioning system, the image feature to generate a list of possible options for a word for the caption and compute a policy network score for each possible option in the list of options;

ranking, by the policy network of the image captioning system, the list of possible options for the word of the caption based on the policy network score for each possible option in the list of options;

analyzing, by a value network of the image captioning system, the image feature and the list of possible options generated by the policy network to generate a value network score for each possible option in the list of options;

ranking, by the value network, the list of options for the word of the caption based on the value network score; and

selecting, by the image captioning system a word for the caption based on the ranking of the list of options by the policy network and the ranking of the list of options by the value network.

2. The method of claim 1, wherein the selected word has the highest combined score of the policy network score and the value network score.

3. The method of claim 1, wherein the value network score is generated by combining each possible option in the list of options with previously generated words in the caption.

4. The method of claim 1, wherein the policy network is pre-trained using supervised learning with cross entropy loss.

5. The method of claim 4, wherein the value network is pre-trained with mean square loss.

6. The method of claim 5, wherein after the pre-training of the policy network and the pre-training of the value network, the policy network and the value network are trained by deep reinforcement learning.

7. The method of claim 1, wherein the policy network comprises a Convolutional Neural Network (CNN) and a Recurrent Neural Network (RNN).

8. The method of claim 7, wherein the policy network generates the list of possible options for the word by encoding visual information of the image feature using the CNN and inputting the encoded visual information into the RNN.

9. The method of claim 1, wherein the value network comprises a CNN, a RNN, and a Multilayer Perceptron (MLP).

10. The method of claim 9, wherein the value network score for each possible option in the list of options is generated by the value network by encoding visual information of the image feature using the CNN, encoding semantic information of a partially generated sentence using the RNN, and regressing a scalar reward from a concatenated visual and semantic feature vector based on the encoded visual information and semantic information.

11. The method of claim 1, wherein selecting the word for the caption further comprises utilizing a lookahead beam search to correct errors using a later word context.

12. An image captioning system comprising:

a memory that stores instructions; and

one or more processors configured by the instructions to perform operations comprising;

extracting, by an image captioning system, an image feature from an image; and

analyzing the image feature to generate a caption describing the image feature by performing the following operations until the caption describing the image feature is complete;

analyzing, by a policy network of the image captioning system, the image feature to generate a list of possible options for a word for the caption and compute a policy network score for each possible option in the list of options;

ranking, by the policy network of the image captioning system, the list of possible options for the word of the caption based on the policy network score for each possible option in the list of options;

analyzing, by a value network of the image captioning system, the image feature and the list of possible options generated by the policy network to generate a value network score for each possible option in the list of options;

ranking, by the value network, the list of options for the word of the caption based on the value network score; and

selecting, by the image captioning system, a word for the caption based on the ranking of the list of options by the policy network and the ranking of the list of options by the value network.

13. The image captioning system of claim 12, wherein the selected word has the highest combined score of the policy network score and the value network score.

14. The image captioning system of claim 12, wherein the value network score is generated by combining each possible option in the list of options with previously generated words in the caption.

15. The image captioning system of claim 12, wherein the policy network is pre-trained using supervised learning with cross entropy loss and wherein the value network is pre-trained with mean square loss.

16. The image captioning system of claim 15, wherein after the pre-training of the policy network and the pre-training of the value network, the policy network and the value network are trained by deep reinforcement learning.

17. The image captioning system of claim 12, wherein the policy network comprises a Convolutional Neural Network (CNN) and a Recurrent Neural Network (RNN) and wherein the policy network generates the list of possible options for the word by encoding visual information of the image feature using the CNN and inputting the encoded visual information into the RNN.

18. The image captioning system of claim 12, wherein the value network comprises a CNN, a RNN, and a Multilayer Perceptron (MLP).

19. The image captioning system of claim 18, wherein the value network score for each possible option in the list of options is generated by the value network by encoding visual information of the image feature using the CNN, encoding semantic information of a partially generated sentence using the RNN, and regressing a scalar reward from a concatenated visual and semantic feature vector based on the encoded visual information and semantic information.

20. A non-transitory computer-readable medium comprising instructions stored thereon that are executable by at least one processor to cause a computing device to perform operations comprising:

extracting, by an image captioning system, an image feature from an image; and

analyzing the image feature to generate a caption describing the image feature by performing the following operations until the caption describing the image feature is complete;

analyzing, by a policy network of the image captioning system, the image feature to generate a list of possible options for a word for the caption and compute a policy network score for each possible option in the list of options;

ranking, by the policy network of the image captioning system, the list of possible options for the word of the caption based on the policy network score for each possible option in the list of options;

analyzing, by a value network of the image captioning system, the image feature and the list of possible options generated by the policy network to generate a value network score for each possible option in the list of options;

ranking, by the value network, the list of options for the word of the caption based on the value network score; and

selecting, by the image captioning system, a word for the caption based on the ranking of the list of options by the policy network and the ranking of the list of options by the value network.