The concept of action recognition in smart security heavily relies on deep learning and artificial intelligence to make predictions about actions of humans. To draw appropriate conclusions from these hypotheses, a large amount of information is required. The data in question are often a video feed, and there is a direct relationship between increased data volume and more-precise decision-making. We seek to determine how far a static video can be compressed before the neural network’s capacity to predict the action in the video is lost. To find this, videos are compressed by lowering the bitrate using FFMPEG. In parallel, a convolutional neural network model is trained to recognise action in the videos and is tested on the compressed videos until the neural network fails to predict the action observed in the videos. The results reveal that bitrate compression has no linear relationship with neural network performance.

Semantic communications focus on transmitting information that encapsulates meaning, enabling both machines and humans to understand the intended message with greater accuracy. Unlike traditional communication systems, which send data without considering its semantic value, this approach prioritises the content's meaning and requires a novel metric to gauge semantic quality. Our framework integrates a specialised Vision Transformer (ViT) for semantic segmentation, named SemExT, at the transmission end and a pre-trained Generative Adversarial Network (GAN) for image reconstruction at the receiving end. The system's effectiveness is evaluated by comparing the semantic content of the reconstructed image with the original, using Deceptron2, an advanced object detection model. This comparison establishes a new metric for assessing the quality of semantic transmission. Empirical evidence shows that the semantic quality metric ranges from 90% to 100% for images containing fewer objects and 80% to 98% for those with more objects. In comparison, an autoencoder-based communication system exhibits a range of 80% to 100% for simpler images and 75% to 95% for more complex ones. These findings highlight the robustness of our proposed metric across different semantic communication frameworks, contributing to the advancement of semantic information transmission and setting a foundation for future research in this field.