Various research video demos with links to available open access manuscripts, open source software and datasets.
360-degree Single Camera Bird-Eye View Mapping
Issue: Conventional approaches to Bird's-Eye View mapping require an on-vehicle multiple camera setup.
Approach: reducing on-vehicle hardware complexity through the use of a single 360-degree camera instead of multiple perspective cameras.
Application: a novel spherical camera autonomous driving dataset equipped with a high-resolution 128- channel 3D LiDAR and a RTK-refined GNSS/INS system, along with a benchmark architecture designed to generate Bird-EyeView (BEV) maps using only a single spherical camera.
Within our benchmark architecture, we propose a novel spherical-image-to-BEV module that leverages spherical imagery and a refined sampling strategy to project features from 2D to 3D.
1 result2025 | ||
| [e25bev] |  | Dur360BEV: A Real-world 360-degree Single Camera Dataset and Benchmark for Bird-Eye View Mapping in Autonomous Driving , In Proc. Int. Conf. on Robotics and Automation, IEEE, pp. 3737-3744, 2025.Keywords: autonomous driving, BEV, spherical camera, 360 degree camera, single camera BEV, surround view, vehicle perception. |
Object-Wise Anomaly Detection in Security X-ray Imagery
Issue: Automatic Prohibited Item Detection Systems (APIDS) based on supervised object detection approaches face additional challenges in the postal mail screening domain.
Approach: A framework leveraging open-world object detection and semi-supervised anomaly detection as a conduit to effective screening in this context.
Application: Specifically considering the context of postal screening, experimental results demonstrate the high recall (77.76%) and accuracy (75.93%) with low false positive rates (1.98%).
We use a cross-modal homography to map complex detected object shapes from the X-ray image to a corresponding colour image of the mail item on the postal conveyor belt, to allow identification of mail items for subeqeuent isolation.
2 results2025 | ||
| [gaus25anomaly] |  | Semi-supervised Object-Wise Anomaly Detection for Firearm and Firearm Component Detection in X-ray Security Imagery , In Proc. Computer Vision Pattern Recognition Workshops, IEEE/CVF, pp. 4004-4014, 2025.Keywords: firearm detection, prohibited item detection, APIDS, anomaly, out-of-distribution, OOD, OWOD, open-world object detection. |
2024 | ||
| [isaac24ssos] |  | Towards Open-World Object-based Anomaly Detection via Self-Supervised Outlier Synthesis , In Proc. European Conference on Computer Vision, Springer, pp. 196-214, 2024.Keywords: x-ray, thermal, anomaly detection, open world object detection, open-set anomaly detection, object-wise anomaly detection. |
Predicting Robot Operator Intention from Brain Signals
Issue: Brain–computer interfaces (BCI) provide a hands-free control modality for mobile robotics, yet decoding user intent during real-world navigation remains challenging.
Approach: an all-terrain robot platform was remotely operated by 12 participants who navigated a predefined outdoor route using a joystick and electroencephalogram (EEG) signals were recorded with a 16-channel BCI headset.
Application: multiple deep learning models were benchmarked for the task of intent classification, across the Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Transformer architectural families.
By combining real-world robotic control with multi-horizon EEG intention decoding, this study introduces a reproducible benchmark and reveals key design insights for predictive, deep leanrning based BCI systems.
1 result2026 | ||
| [alosaimi26eeg] |  | EEG-Driven Intention Decoding: Offline Deep Learning Benchmarking on a Robotic Rover , In Proc. Int. Conf. on Robotics and Automation, IEEE, 2026. (to appear)Keywords: EEG, BCI, brain computer interface, cnn, neural networks, convolutional neural networks, deep learning, dry-eeg, robot guidance, rover. |
Region-based Anomaly Detection in Thermal Imagery
Issue: the key challenge of Anomaly detection is namely the determination of the normal from the abnormal when operational data availability is highly biased towards one class (normal) due to both insufficient sample size, and inadequate distribution coverage for the other class (abnormal).
Approach: we propose the dual use of both visual appearance and localized motion characteristics, derived from optic flow, applied on a per-region basis to facilitate object-wise anomaly detection within this context.
Application: by leveraging established object localization techniques, optic flow is extracted from each object region and combined with appearance in the far infrared (thermal) band to give a 3-channel spatiotemporal tensor representation for each object (1 × thermal - spatial appearance; 2 × op- tic flow magnitude as x and y components - temporal motion).
This formulation is used as the basis for training contemporary semi-supervised anomaly detection approaches in a region-based manner such that anomalous objects can be detected as a combination of appearance and/or motion within the scene.
1 result2023 | ||
| [gaus23region] |  | Region-based Appearance and Flow Characteristics for Anomaly Detection in Infrared Surveillance Imagery , In Proc. Conf. Computer Vision and Pattern Recognition Workshops, IEEE/CVF, pp. 2995-3005, 2023.Keywords: anomaly detection, optic flow, thermal imagery, visual surveillance, R-CNN, convolutional neural network, deep learning. |