Boris Bačić

Munish Rathee, Boris Bačić, Maryam Doborjeh

This paper presents the SIFT-SNN framework, a low-latency neuromorphic signal-processing pipeline for real-time detection of structural anomalies in transport infrastructure. The proposed approach integrates Scale-Invariant Feature Transform (SIFT) for spatial feature encoding with a latency-driven spike conversion layer and a Leaky Integrate-and-Fire (LIF) Spiking Neural Network (SNN) for classification. The Auckland Harbour Bridge dataset is recorded under various weather and lighting conditions, comprising 6,000 labelled frames that include both real and synthetically augmented unsafe cases. The presented system achieves a classification accuracy of 92.3% (+- 0.8%) with a per-frame inference time of 9.5 ms. Achieved sub-10 millisecond latency, combined with sparse spike activity (8.1%), enables real-time, low-power edge deployment. Unlike conventional CNN-based approaches, the hybrid SIFT-SNN pipeline explicitly preserves spatial feature grounding, enhances interpretability, supports transparent decision-making, and operates efficiently on embedded hardware. Although synthetic augmentation improved robustness, generalisation to unseen field conditions remains to be validated. The SIFT-SNN framework is validated through a working prototype deployed on a consumer-grade system and framed as a generalisable case study in structural safety monitoring for movable concrete barriers, which, as a traffic flow-control infrastructure, is deployed in over 20 cities worldwide.

Boris Bačić, Claudiu Vasile, Chengwei Feng et al.

The purpose of this paper is to contribute towards the near-future privacy-preserving big data analytical healthcare platforms, capable of processing streamed or uploaded timeseries data or videos from patients. The experimental work includes a real-life knee rehabilitation video dataset capturing a set of exercises from simple and personalised to more general and challenging movements aimed for returning to sport. To convert video from mobile into privacy-preserving diagnostic timeseries data, we employed Google MediaPipe pose estimation. The developed proof-of-concept algorithms can augment knee exercise videos by overlaying the patient with stick figure elements while updating generated timeseries plot with knee angle estimation streamed as CSV file format. For patients and physiotherapists, video with side-to-side timeseries visually indicating potential issues such as excessive knee flexion or unstable knee movements or stick figure overlay errors is possible by setting a-priori knee-angle parameters. To address adherence to rehabilitation programme and quantify exercise sets and repetitions, our adaptive algorithm can correctly identify (91.67%-100%) of all exercises from side- and front-view videos. Transparent algorithm design for adaptive visual analysis of various knee exercise patterns contributes towards the interpretable AI and will inform near-future privacy-preserving, non-vendor locking, open-source developments for both end-user computing devices and as on-premises non-proprietary cloud platforms that can be deployed within the national healthcare system.

Martin Štufi, Boris Bačić

There is a growing trend for enterprise-level Internet of Things (IoT) applications requiring real-time horizontally scalable data processing platforms. Real-time processing platforms receiving data streams from sensor networks (e.g., autonomous and connected vehicles, smart security for businesses and homes, smartwatches, fitness trackers, and other wearables) require distributed MQTT brokers. This case study presents an IoT data streaming testbed platform prepared for the Czech Post. The presented platform has met the throughput requirement of 2 million messages per 24 hours (comprising SMS and emails). The tested MQTT broker runs on a single virtual node of a horizontally scalable testbed platform. Soon the Czech Post will modernise its eServices to increase package deliveries aligned with eCommerce and eGovernment demands. The presented testbed platform fulfils all requirements, and it is also capable of processing thousands of messages per second. The presented platform and concepts are transferable to healthcare systems, transport operations, the automotive industry, and other domains such as smart cities.

Boris Bačić, Jason Zhang

The primary focus of this paper is to produce a proof of concept for extracting drowsiness information from videos to help elderly living on their own. To quantify yawning, eyelid and head movement over time, we extracted 3000 images from captured videos for training and testing of deep learning models integrated with OpenCV library. The achieved classification accuracy for eyelid and mouth open/close status were between 94.3%-97.2%. Visual inspection of head movement from videos with generated 3D coordinate overlays, indicated clear spatiotemporal patterns in collected data (yaw, roll and pitch). Extraction methodology of the drowsiness information as timeseries is applicable to other contexts including support for prior work in privacy-preserving augmented coaching, sport rehabilitation, and integration with big data platform in healthcare.

Boris Bačić

Current exergaming sensors and inertial systems attached to sports equipment or the human body can provide quantitative information about the movement or impact e.g. with the ball. However, the scope of these technologies is not to qualitatively assess sports technique at a personalised level, similar to a coach during training or replay analysis. The aim of this paper is to demonstrate a novel approach to automate identification of tennis swings executed with erroneous technique without recorded ball impact. The presented spatiotemporal transformations relying on motion gradient vector flow and polynomial regression with RBF classifier, can identify previously unseen erroneous swings (84.5-94.6%). The presented solution is able to learn from a small dataset and capture two subjective swing-technique assessment criteria from a coach. Personalised and flexible assessment criteria required for players of diverse skill levels and various coaching scenarios were demonstrated by assigning different labelling criteria for identifying similar spatiotemporal patterns of tennis swings.

Boris Bačić, Patria Hume

Coaching technology, wearables and exergames can provide quantitative feedback based on measured activity, but there is little evidence of qualitative feedback to aid technique improvement. To achieve personalised qualitative feedback, we demonstrated a proof-of-concept prototype combining kinesiology and computational intelligence that could help improving tennis swing technique utilising three-dimensional tennis motion data acquired from multi-camera video. Expert data labelling relied on virtual 3D stick figure replay. Diverse assessment criteria for novice to intermediate skill levels and configurable coaching scenarios matched with a variety of tennis swings (22 backhands and 21 forehands), included good technique and common errors. A set of selected coaching rules was transferred to adaptive assessment modules able to learn from data, evolve their internal structures and produce autonomous personalised feedback including verbal cues over virtual camera 3D replay and an end-of-session progress report. The prototype demonstrated autonomous assessment on future data based on learning from prior examples, aligned with skill level, flexible coaching scenarios and coaching rules. The generated intuitive diagnostic feedback consisted of elements of safety and performance for tennis swing technique, where each swing sample was compared with the expert. For safety aspects of the relative swing width, the prototype showed improved assessment ...