Scene recognition method, training method, device, electronic equipment and program product

Abstract

Embodiments of the present invention provide a scene recognition method, training method, device, electronic device, and program product. The method includes: acquiring a color image and a depth image of a scene to be recognized; using a feature extraction algorithm to characterize the color image and the depth image respectively Extract and obtain the first global feature and the first local feature corresponding to the color image, and the second global feature and the second local feature corresponding to the depth image; the first global feature, the second global feature, the first local feature and the second The local features are fused to obtain the multimodal features of the scene to be recognized; the scene recognition is performed on the multimodal features of the scene to be recognized to obtain the recognition result of the scene to be recognized. On the basis of extracting the global features of the color image and the depth image, the first local feature and the second local feature of the color image and the depth image are also extracted respectively, and the above-mentioned local features and global features are fused, thereby improving scene recognition. accuracy.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular, to a scene recognition method, a training method, an apparatus, an electronic device and a program product. Background technique [0002] Scene recognition is an important branch of computer vision technology and has been widely used in various fields. With the outstanding performance of Convolutional Neural Network (CNN) in image classification tasks, more and more scene recognition is based on convolutional neural networks. There are a lot of irrelevant information and interference information in the actual indoor scene, so the research on indoor scene recognition is of great significance. [0003] In the prior art, the acquired color image and depth image of the scene to be recognized are generally extracted based on the convolutional neural network model, respectively, and the fusion of the above two images is realized by cascade processing or element-level addit...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a scene recognition method, training method, device, electronic equipment, and program product to solve the problem of ignoring the local features of the color image and the depth image of the scene to be recognized and the semantics between the two images in the existing method Differences, leading to technical problems with low accuracy of scene recognition

Images