In this paper, we propose an access control method with a secret key for object detection models for the first time so that unauthorized users without a secret key cannot benefit from the performance of trained models. The method enables us not only to provide a high detection performance to authorized users but to also degrade the performance for unauthorized users. The use of transformed images was proposed for the access control of image classification models, but these images cannot be used for object detection models due to performance degradation. Accordingly, in this paper, selected feature maps are encrypted with a secret key for training and testing models, instead of input images. In an experiment, the protected models allowed authorized users to obtain almost the same performance as that of non-protected models but also with robustness against unauthorized access without a key.

With massive utilization of video in every aspect of our daily lives, managing videos is crucial and demanding. The rich literature of data embedding has proven its viability in managing as well as enriching videos and other multimedia contents, but conventional methods are designed to operate in the media/compression layer. In this work, the synchronization between the audio-video and subtitle tracks within an MP4 format container is manipulated to insert data. Specifically, the data are derived from the statistics of the audio samples and video frames, and it serves as the authentication data for verification purpose. When needed, the inserted authentication data can be extracted and compared against the information computed from the received audio samples and video frames. The proposed method is lightweight because simple statistics, i.e., ‘0’ and ‘1’ at the bit stream level, are treated as the authentication data. Furthermore, unlike conventional MP4 container format-based data insertion technique, the bit stream size remains unchanged before and after data insertion using the proposed method. The proposed authentication method can also be deployed for joint utilization with any existing authentication technique for audio / video as long as these media can be multiplexed into a single bit stream and contained within an MP4 container. Experiments are carried out to verify the basic functionality of the proposed technique as an authentication method.

In this paper, we propose a combined use of transformed images and vision transformer (ViT) models transformed with a secret key. We show for the first time that models trained with plain images can be directly transformed to models trained with encrypted images on the basis of the ViT architecture, and the performance of the transformed models is the same as models trained with plain images when using test images encrypted with the key. In addition, the proposed scheme does not require any specially prepared data for training models or network modification, so it also allows us to easily update the secret key. In an experiment, the effectiveness of the proposed scheme is evaluated in terms of performance degradation and model protection performance in an image classification task on the CIFAR-10 dataset.