Skip to main content

Research Repository

Advanced Search

All Outputs (36)

Object landmark discovery through unsupervised adaptation (2019)
Journal Article
Sanchez, E., & Tzimiropoulos, G. (2019). Object landmark discovery through unsupervised adaptation. Advances in Neural Information Processing Systems, 32,

This paper proposes a method to ease the unsupervised learning of object landmark detectors. Similarly to previous methods, our approach is fully unsupervised in a sense that it does not require or make any use of annotated landmarks for the target o... Read More about Object landmark discovery through unsupervised adaptation.

Rapid tracking of extrinsic projector parameters in fringe projection using machine learning (2018)
Journal Article
Stavroulakis, P., Chen, S., Delorme, C., Bointon, P., Tzimiropoulos, G., & Leach, R. (2019). Rapid tracking of extrinsic projector parameters in fringe projection using machine learning. Optics and Lasers in Engineering, 114, 7-14. https://doi.org/10.1016/j.optlaseng.2018.08.018

In this work, we propose to enable the angular re-orientation of a projector within a fringe projection system in real-time without the need for re-calibrating the system. The estimation of the extrinsic orientation parameters of the projector is per... Read More about Rapid tracking of extrinsic projector parameters in fringe projection using machine learning.

Zero-Shot Keyword Spotting for Visual Speech Recognition In-the-wild (2018)
Conference Proceeding
Tzimiropoulos, Y., & Stafylakis, T. (2018). Zero-Shot Keyword Spotting for Visual Speech Recognition In-the-wild. In Computer Vision – ECCV 2018 (536-552). https://doi.org/10.1007/978-3-030-01225-0_32

Visual keyword spotting (KWS) is the problem of estimating whether a text query occurs in a given recording using only video information. This paper focuses on visual KWS for words unseen during training, a real-world, practical setting which so far... Read More about Zero-Shot Keyword Spotting for Visual Speech Recognition In-the-wild.

To Learn Image Super-Resolution, Use a GAN to Learn How to Do Image Degradation First (2018)
Conference Proceeding
Bulat, A., Yang, J., & Tzimiropoulos, G. (2018). To Learn Image Super-Resolution, Use a GAN to Learn How to Do Image Degradation First. In Computer Vision – ECCV 2018: 15th European Conference Munich, Germany, September 8–14, 2018 Proceedings, Part VI (187-202). https://doi.org/10.1007/978-3-030-01231-1_12

© Springer Nature Switzerland AG 2018. This paper is on image and face super-resolution. The vast majority of prior work for this problem focus on how to increase the resolution of low-resolution images which are artificially generated by simple bili... Read More about To Learn Image Super-Resolution, Use a GAN to Learn How to Do Image Degradation First.

Hierarchical binary CNNs for landmark localization with limited resources (2018)
Journal Article
Bulat, A., & Tzimiropoulos, G. (2020). Hierarchical binary CNNs for landmark localization with limited resources. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2), 343 - 356. https://doi.org/10.1109/tpami.2018.2866051

Our goal is to design architectures that retain the groundbreaking performance of Convolutional Neural Networks (CNNs) for landmark localization and at the same time are lightweight, compact and suitable for applications with limited computational re... Read More about Hierarchical binary CNNs for landmark localization with limited resources.

Novel monitoring systems to obtain dairy cattle phenotypes associated with sustainable production (2018)
Journal Article
Bell, M. J., & Tzimiropoulos, G. (2018). Novel monitoring systems to obtain dairy cattle phenotypes associated with sustainable production. Frontiers in Sustainable Food Systems, 2, Article 31. https://doi.org/10.3389/fsufs.2018.00031

Improvements in production efficiencies and profitability of products from cattle are of great interest to farmers. Furthermore, improvements in production efficiencies associated with feed utilization and fitness traits have also been shown to reduc... Read More about Novel monitoring systems to obtain dairy cattle phenotypes associated with sustainable production.

End-to-end audiovisual speech recognition (2018)
Conference Proceeding
Petridis, S., Stafylakis, T., Ma, P., Cai, F., Tzimiropoulos, G., & Pantic, M. (2018). End-to-end audiovisual speech recognition.

Several end-to-end deep learning approaches have been recently presented which extract either audio or visual features from the input images or audio signals and perform speech recognition. However, research on end-to-end audiovisual models is very l... Read More about End-to-end audiovisual speech recognition.

Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression (2017)
Conference Proceeding
Jackson, A. S., Bulat, A., Argyriou, V., & Tzimiropoulos, G. (2017). Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression. In Proceedings - 2017 IEEE International Conference on Computer Vision (1031-1039). https://doi.org/10.1109/ICCV.2017.117

3D face reconstruction is a fundamental Computer Vision problem of extraordinary difficulty. Current systems often assume the availability of multiple facial images (sometimes from the same subject) as input, and must address a number of methodologic... Read More about Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression.

Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources (2017)
Conference Proceeding
Bulat, A., & Tzimiropoulos, G. (2017). Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources. In Proceedings - 2017 IEEE International Conference on Computer Vision (ICCV 2017) (3726 - 3734). https://doi.org/10.1109/ICCV.2017.400

Our goal is to design architectures that retain the groundbreaking performance of CNNs for landmark localization and at the same time are lightweight, compact and suitable for applications with limited computational resources. To this end, we make th... Read More about Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources.

How Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks) (2017)
Conference Proceeding
Bulat, A., & Tzimiropoulos, G. (2017). How Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks). In 2017 IEEE International Conference on Computer Vision (ICCV 2017) (1021-1030). https://doi.org/10.1109/ICCV.2017.116

This paper investigates how far a very deep neural network is from attaining close to saturating performance on existing 2D and 3D face alignment datasets. To this end, we make the following 5 contributions: (a) we construct, for the first time, a ve... Read More about How Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks).

Synergy between face alignment and tracking via Discriminative Global Consensus Optimization (2017)
Conference Proceeding
Khan, M. H., McDonagh, J., & Tzimiropoulos, G. (2017). Synergy between face alignment and tracking via Discriminative Global Consensus Optimization.

An open question in facial landmark localization in video is whether one should perform tracking or tracking-by-detection (i.e. face alignment). Tracking produces fittings of high accuracy but is prone to drifting. Tracking-by-detection is drift-free... Read More about Synergy between face alignment and tracking via Discriminative Global Consensus Optimization.

Deep machine learning provides state-of-the-art performance in image-based plant phenotyping (2017)
Journal Article
Pound, M. P., Atkinson, J. A., Townsend, A. J., Wilson, M. H., Griffiths, M., Jackson, A. S., …French, A. P. (2017). Deep machine learning provides state-of-the-art performance in image-based plant phenotyping. GigaScience, 6(10), Article gix083. https://doi.org/10.1093/gigascience/gix083

© The Author 2017. In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, h... Read More about Deep machine learning provides state-of-the-art performance in image-based plant phenotyping.

Fast and exact Newton and bidirectional fitting of Active Appearance Models (2016)
Journal Article
Kossaifi, J., Tzimiropoulos, G., & Pantic, M. (2017). Fast and exact Newton and bidirectional fitting of Active Appearance Models. IEEE Transactions on Image Processing, 26(2), 1040 - 1053. https://doi.org/10.1109/TIP.2016.2642828

Active Appearance Models (AAMs) are generative models of shape and appearance that have proven very attractive for their ability to handle wide changes in illumination, pose and occlusion when trained in the wild, while not requiring large training d... Read More about Fast and exact Newton and bidirectional fitting of Active Appearance Models.

Frequency domain subpixel registration using HOG phase correlation (2016)
Journal Article
Argyriou, V., & Tzimiropoulos, G. (in press). Frequency domain subpixel registration using HOG phase correlation. Computer Vision and Image Understanding, https://doi.org/10.1016/j.cviu.2016.10.019

We present a novel frequency-domain image registration technique, which employs histograms of oriented gradients providing subpixel estimates. Our method involves image filtering using dense Histogram of Oriented Gradients (HOG), which provides an ad... Read More about Frequency domain subpixel registration using HOG phase correlation.

Fast algorithms for fitting active appearance models to unconstrained images (2016)
Journal Article
Tzimiropoulos, G., & Pantic, M. (in press). Fast algorithms for fitting active appearance models to unconstrained images. International Journal of Computer Vision, https://doi.org/10.1007/s11263-016-0950-1

Fitting algorithms for Active Appearance Models (AAMs) are usually considered to be robust but slow or fast but less able to generalize well to unseen variations. In this paper, we look into AAM fitting algorithms and make the following orthogonal co... Read More about Fast algorithms for fitting active appearance models to unconstrained images.

Convolutional aggregation of local evidence for large pose face alignment (2016)
Conference Proceeding
Bulat, A., & Tzimiropoulos, G. (2016). Convolutional aggregation of local evidence for large pose face alignment.

Methods for unconstrained face alignment must satisfy two requirements: they must not rely on accurate initialisation/face detection and they should perform equally well for the whole spectrum of facial poses. To the best of our knowledge, there are... Read More about Convolutional aggregation of local evidence for large pose face alignment.

Human Pose Estimation via Convolutional Part Heatmap Regression (2016)
Conference Proceeding
Bulat, A., & Tzimiropoulos, G. (2016). Human Pose Estimation via Convolutional Part Heatmap Regression. In Computer Vision – ECCV 2016 (717-732). https://doi.org/10.1007/978-3-319-46478-7_44

This paper is on human pose estimation using Convolutional Neural Networks. Our main contribution is a CNN cascaded architecture specifically designed for learning part relationships and spatial context, and robustly inferring pose even for the case... Read More about Human Pose Estimation via Convolutional Part Heatmap Regression.

Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques (2016)
Journal Article
Dupre, R., Argyriou, V., Tzimiropoulos, G., & Greenhill, D. (2016). Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques. Information Sciences, 372, https://doi.org/10.1016/j.ins.2016.08.075

In this paper, the notion of risk analysis within 3D scenes using vision based techniques is introduced. In particular the problem of risk estimation of indoor environments at the scene and object level is considered, with applications in domestic ro... Read More about Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques.