Skip to main content

Research Repository

Advanced Search

Outputs (204)

An Optimised Deep Neural Network Approach for Forest Trail Navigation for UAV Operation within the Forest Canopy (2017)
Presentation / Conference Contribution
Maciel-Pearson, B., & Breckon, T. (2017, December). An Optimised Deep Neural Network Approach for Forest Trail Navigation for UAV Operation within the Forest Canopy. Presented at The UK-RAS Network Conference on Robotics and Autonomous Systems: robots working for and among us., Bristol, England

Autonomous flight within a forest canopy represents a key challenge for generalised scene understanding on-board a future Unmanned Aerial Vehicle (UAV) platform. Here we present an approach for automatic trail navigation within such an environment th... Read More about An Optimised Deep Neural Network Approach for Forest Trail Navigation for UAV Operation within the Forest Canopy.

Clustering in pursuit of temporal correlation for human motion segmentation (2017)
Journal Article
Qian, C., Breckon, T., & Xu, Z. (2018). Clustering in pursuit of temporal correlation for human motion segmentation. Multimedia Tools and Applications, 77(15), 19615-19631. https://doi.org/10.1007/s11042-017-5408-0

Temporal correlation is an important property of the video sequence. However, most methods only accomplish the clustering of frames via the measurement of similarity between frame pair, and the temporal correlation among frames is rarely taken into a... Read More about Clustering in pursuit of temporal correlation for human motion segmentation.

Face Recognition via Deep Sparse Graph Neural Networks (2017)
Presentation / Conference Contribution
Wu, R., Kamata, S., & Breckon, T. (2017, September). Face Recognition via Deep Sparse Graph Neural Networks. Presented at 28th British Machine Vision Conference (BMVC) 2017., London, UK

DepthComp: Real-time Depth Image Completion Based on Prior Semantic Scene Segmentation (2017)
Presentation / Conference Contribution
Atapour-Abarghouei, A., & Breckon, T. (2017, September). DepthComp: Real-time Depth Image Completion Based on Prior Semantic Scene Segmentation. Presented at 28th British Machine Vision Conference (BMVC) 2017, London, UK

We address plausible hole filling in depth images in a computationally lightweight methodology that leverages recent advances in semantic scene segmentation. Firstly, we perform such segmentation over a co-registered color image, commonly available f... Read More about DepthComp: Real-time Depth Image Completion Based on Prior Semantic Scene Segmentation.

Back to Butterworth - a Fourier Basis for 3D Surface Relief Hole Filling within RGB-D Imagery (2016)
Presentation / Conference Contribution
Atapour-Abarghouei, A., de La Garanderie, G. P., & Breckon, T. P. (2016, December). Back to Butterworth - a Fourier Basis for 3D Surface Relief Hole Filling within RGB-D Imagery. Presented at 2016 23rd International Conference on Pattern Recognition (ICPR), Cancun

We address the problem of hole filling in RGB-D (color and depth) images, obtained from either active or stereo based sensing, for the purposes of object removal and missing depth estimation. This is performed independently on the low frequency depth... Read More about Back to Butterworth - a Fourier Basis for 3D Surface Relief Hole Filling within RGB-D Imagery.

Real-time Classification of Vehicle Types within Infra-red Imagery (2016)
Presentation / Conference Contribution
Kundegorski, M., Akcay, S., Payen de La Garanderie, G., Breckon, T., & Stokes, R. (2016, November). Real-time Classification of Vehicle Types within Infra-red Imagery. Presented at Optics and Photonics for Counterterrorism, Crime Fighting and Defence XII, Edinburgh, United Kingdom

Real-time classification of vehicles into sub-category types poses a significant challenge within infra-red imagery due to the high levels of intra-class variation in thermal vehicle signatures caused by aspects of design, current operating duration... Read More about Real-time Classification of Vehicle Types within Infra-red Imagery.

Noise Robust Image Edge Detection based upon the Automatic Anisotropic Gaussian Kernels (2016)
Journal Article
Zhang, W., Zhao, Y., Breckon, T., & Chen, L. (2016). Noise Robust Image Edge Detection based upon the Automatic Anisotropic Gaussian Kernels. Pattern Recognition, 63(8), 193-205. https://doi.org/10.1016/j.patcog.2016.10.008

This paper presents a novel noise robust edge detector based upon the automatic anisotropic Gaussian kernels (ANGKs), which also addresses the current problem that the seminal Canny edge detector may miss some obvious crossing edge details. Firstly,... Read More about Noise Robust Image Edge Detection based upon the Automatic Anisotropic Gaussian Kernels.

From On-Road to Off: Transfer Learning within a Deep Convolutional Neural Network for Segmentation and Classification of Off-Road Scenes (2016)
Presentation / Conference Contribution
Holder, C., Breckon, T., & Wei, X. (2016, December). From On-Road to Off: Transfer Learning within a Deep Convolutional Neural Network for Segmentation and Classification of Off-Road Scenes. Presented at European Conference on Computer Vision Workshops., Amsterdam, The Netherlands

Real-time road-scene understanding is a challenging computer vision task with recent advances in convolutional neural networks (CNN) achieving results that notably surpass prior traditional feature driven approaches. Here, we take an existing CNN arc... Read More about From On-Road to Off: Transfer Learning within a Deep Convolutional Neural Network for Segmentation and Classification of Off-Road Scenes.

Constant-time Bilateral Filter using Spectral Decomposition (2016)
Presentation / Conference Contribution
Sugimoto, K., Breckon, T., & Kamata, S. (2016, September). Constant-time Bilateral Filter using Spectral Decomposition. Presented at 2016 IEEE International Conference on Image Processing (ICIP)., Phoenix, AZ, USA

This paper presents an efficient constant-time bilateral filter where constant-time means that computational complexity is independent of filter window size. Many state-of-the-art constant-time methods approximate the original bilateral filter by an... Read More about Constant-time Bilateral Filter using Spectral Decomposition.