Selected Videos

Hierarchical Probabilistic Human Body Pose and Shape Prediction

A presentation video for a CVPR 2023 (Vancouver) paper on HuManiFlow: Ancestor-Conditioned Flows on SO(3) for Human Pose and Shape Prediction. See [37] for more detail.

IMP: Iterative Matching and Pose Estimation with Adaptive Pooling

A presentation video for a CVPR 2023 (Vancouver) paper on IMP: Iterative Matching and Pose Estimation With Adaptive Pooling. See [35] for more detail.

Iterative Refinement of Single-View Depth using Surface Normal and its Uncertainty

A demo video showing our iterative 3D shape estimation approach described in Iterative Refinement of Single-View Depth using Surface Normal and its Uncertainty paper at BMVC 2022 (London). See [33] for more detail.

SPARC: Sparse Render-and-Compare for CAD Model Alignment from a single Image

A demo video showing our 3D pose estimation approach based on sparse render and compare paradigm, described in SPARC: Sparse Render-and-Compare for CAD Model Alignment from a Single Image paper at BMVC 2022 (London). See [32] for more detail.

3D Shape and Pose Estimation in the Wild

A demo video showing our 3D Shape and Pose Estimation of human body approach described in Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild paper at ICCV 2021 (Virtual). See [24] for more detail.

Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild

This is a conference presentation for Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild paper [24] at ICCV 2021 (Virtual).

Large Scale Joint Semantic Re-Localisation and Scene Understanding via Globally Unique Instance Coordinate Regression

A supplementary video demonstrating our large-scale semantic re-localisation approach described in Large Scale Joint Semantic Re-Localisation and Scene Understanding via Globally Unique Instance Coordinate Regression paper at BMVC 2019 (Cardiff). See [16] for more detail.

Large Scale Semantic Relocalisation under Changes in Environment

This video illustrates simultaneous large scale joint semantic re-localisation and scene understanding results on artificial city - SceneCity. Top row shows localisation results in a city under mild changes in trajectory and no changes in the environment. The bottom row shows localisation results under significant changes in the map (20% of the buildings are removed). See more detail in Large Scale Joint Semantic Re-Localisation and Scene Understanding via Globally Unique Instance Coordinate Regression paper [16] at BMVC 2018 (Cardiff).

Semantic Localisation via Globally Unique Instance Segmentation - 360 Video

A supplementary video demonstrating four key stages: video collection, hand labeling and label propagation and localisation of our semantic localisation approach described in Semantic Localisation via Globally Unique Instance Segmentation paper at BMVC 2018 (Newcastle). See [15] for more detail.

Semantic Localisation via Globally Unique Instance Segmentation - Detailed Analysis

This video illustrates simultaneous localisation and surrounding environment recognition results on CamVid-360 dataset. See more detail in Semantic Localisation via Globally Unique Instance Segmentation paper [15] at BMVC 2018 (Newcastle).

Semantic Localisation in an Artificial City under Changes of Environment - Missing Buildings

This video illustrates simultaneous localisation and surrounding environment recognition results on SceneCity (artificial city) dataset. See details in Semantic Localisation via Globally Unique Instance Segmentation paper [15] at BMVC 2018 (Newcastle).

Semantic Localisation in a Highly Repetitive Large Artificial City - more than 800 Buildings

This video illustrates simultaneous localisation and surrounding environment recognition results on SceneCity (artificial city) dataset. See details in Semantic Localisation via Globally Unique Instance Segmentation paper [15] at BMVC 2018 (Newcastle).

Indirect Deep Structured Learning

A supplementary video demonstrating results of Indirect Deep Structured Learning for 3D Human Body Shape and Pose Prediction paper at BMVC 2017. See [13].

Large Scale Augmentation of Semantic Labels in Videos - CamVid

Example video of semantic label propagation from Large Scale Labelled Video Data Augmentation for Semantic Segmentation in Driving Scenarios at the 5th Workshop on Computer Vision for Road Scene Understanding and Autonomous Driving, ICCV 2017. See [14].

Large Scale Augmentation of Semantic Labels in Videos - CityScapes

Example video of instance label propagation from Large Scale Labelled Video Data Augmentation for Semantic Segmentation in Driving Scenarios at the 5th Workshop on Computer Vision for Road Scene Understanding and Autonomous Driving, ICCV 2017. See [14].

Large Scale Augmentation of Instance Labels in Videos - CamVid-Instance

Example video of instance label propagation from Large Scale Labelled Video Data Augmentation for Semantic Segmentation in Driving Scenarios at the 5th Workshop on Computer Vision for Road Scene Understanding and Autonomous Driving, ICCV 2017. See [14].

3D Human Body Shape and Pose Prediction

An early attempt to predict 3D shape and pose in videos using indirect deep structured learning. See [13].

-

All publications



[37] Sengupta, A., Budvytis, I., Cipolla, R.
HuManiFlow: Ancestor-Conditioned Normalising Flows on SO(3) Manifolds for Human Pose and Shape Distribution Estimation [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, June 2023.

[36] Xue, F., Budvytis, I., Cipolla, R.
SFD2: Semantic-guided Feature Detection and Description [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, June 2023.

[35] Xue, F., Budvytis, I., Cipolla, R.
IMP: Iterative Matching and Pose Estimation with Adaptive Pooling [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, June 2023.

[34] Logothetis, F., Mecca, R., Budvytis, I., Cipolla,R.,
A CNN based Approach for the Point-Light Photometric Stereo Problem [pdf]
International Journal of Computer Vision (IJCV), 131(1):101-120, 2023.

[33] Bae, G., Budvytis, I., Cipolla, R.
Irondepth: Iterative Refinement of Single-View Depth using Surface Normal and its Uncertainty [pdf] [oral presentation]
In Proc. British Machine Vision Conference (BMVC), London, November 2022.

[32] Langer, F., Bae, G., Budvytis, I., Cipolla, R.
SPARC: Sparse Render-and-Compare for CAD Model Alignment in a Single RGB Image [pdf]
In Proc. British Machine Vision Conference (BMVC), London, November 2022.

[31] Xue, F., Budvytis, I., Reino, D.O., Cipolla, R.
Efficient Large-Scale Localization by Global Instance Recognition [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, June 2022.

[30] Bae, G., Budvytis, I., Cipolla, R.
Multi-View Depth Estimation by Fusing Single-View Depth Probability with Multi-View Geometry [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, June 2022.

[29] Zhao, C., Budvytis, I., Liwicki, S., Cipolla, R.
Lifted Semantic Graph Embedding for Omnidirectional Place Recognition [pdf]
In Proc. International Conference on 3D Vision (3DV), Virtual, December 2021.

[28] Sengupta, A., Budvytis, I., Cipolla, R.
Probabilistic Estimation of 3D Human Shape and Pose with a Semantic Local Parametric Model [pdf]
In Proc. British Machine Vision Conference (BMVC), Virtual, September 2021.

[27] Langer, F., Budvytis, I., Cipolla, R.
Leveraging Geometry for Shape Estimation from a Single RGB Image [pdf]
In Proc. British Machine Vision Conference (BMVC), Virtual, September 2021.

[26] Mecca, R., Logothetis, F., Budvytis, I., Cipolla, R.
LUCES: A Dataset for Near-Field Point Light Source Photometric Stereo [pdf]
In Proc. British Machine Vision Conference (BMVC), Virtual, September 2021.

[25] Bae, G., Budvytis, I., Cipolla, R.
Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation [pdf] [oral presentation]
In Proc. IEEE International Conference on Computer Vision (ICCV), Virtual, October 2021.

[24] Sengupta, A., Budvytis, I., Cipolla, R.
Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild [pdf][code]
In Proc. IEEE International Conference on Computer Vision (ICCV), Virtual, October 2021.

[23] Logothetis, F., Budvytis, I., Mecca, R., Cipolla, R.
PX-NET: Simple and Efficient Pixel-Wise Training of Photometric Stereo Networks [pdf]
In Proc. IEEE International Conference on Computer Vision (ICCV), Virtual, October 2021.

[22] Sengupta, A., Budvytis, I., Cipolla, R.
Probabilistic 3D Human Shape and Pose Estimation from Multiple Unconstrained Images in the Wild [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Virtual, June 2021.

[21] Vojir, T., Budvytis, I., Cipolla, R.
Efficient Large-Scale Semantic Visual Localization in 2D Maps [pdf][oral presentation]
In Proc. Asian Conference on Computer Vision (ACCV), Virtual Kyoto, November 2020.

[20] Zhang, C., Budvytis, I., Liwicki, S., Cipolla, R.
Rotation Equivariant Orientation Estimation for Omnidirectional Localization [pdf]
In Proc. Asian Conference on Computer Vision (ACCV), Virtual Kyoto, November 2020.

[19] Logothetis, F., Budvytis, I., Mecca, R., Cipolla, R.
A CNN based approach for the near-field photometric stereo problem [pdf][oral presentation] - Best industry paper prize
In Proc. British Machine Vision Conference (BMVC), Virtual, September 2020.

[18] Sengupta, A., Budvytis, I., Cipolla, R.
Synthetic training for accurate 3d human pose and shape estimation in the wild [pdf]
In Proc. British Machine Vision Conference (BMVC), Virtual, September 2020.

[17] Bae, G., Budvytis, I., Yeung, C-K., Cipolla, R.
Deep Multi-view Stereo for Dense 3D Reconstruction from Monocular Endoscopic Video [pdf]
In Proc. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Lima, September 2020

[16] Budvytis*, I., Teichmann*, M., Vojir*, T., Cipolla, R.,
Large Scale Joint Semantic Re-Localisation and Scene Understanding via Globally Unique Instance Coordinate Regression. [pdf]
In Proc. British Machine Vision Conference (BMVC), Cardiff, September 2019
* - indicates equal contribution

[15] Budvytis, I., Sauer, P., Cipolla, R.,
Semantic localisation via globally unique instance segmentation. [pdf][dataset-available-via-email]
In Proc. British Machine Vision Conference (BMVC), Newcastle, September 2018

[14] Budvytis, I., Sauer, P., Roddick, T., Breen, K., Cipolla, R.,
Large scale labelled video data augmentation for semantic segmentation in driving scenarios. [code][pdf][dataset-available-via-email]
In 5th Workshop on Computer Vision for Road Scene Understanding and Autonomous Driving in IEEE International Conference on Computer Vision (ICCV), Venice, October 2017

[13] Tan, J., Budvytis, I., Cipolla, R.,
Indirect deep structured learning for 3D human body shape and pose prediction. [pdf] [oral presentation] [slides-pdf]
In Proc. British Machine Vision Conference (BMVC), London, September 2017

[12] Charles, J., Budvytis, I., Cipolla, R.,
Real-time Factored ConvNets: Extracting the X Factor in Human Parsing. [pdf]
In Proc. British Machine Vision Conference (BMVC), London, September 2017

[11] Badrinarayanan, V., Budvytis, I., Cipolla,R.,
Mixture of Trees Probabilistic Graphical Model for Video Segmentation. [pdf]
International Journal of Computer Vision (IJCV), BMVC12 special issue, December 2013

[10] Budvytis, I., advised by Badrinarayanan V., supervised by Cipolla, R.,
Novel Probabilistic Graphical Models for Semi-Supervised Video Segmentation. [pdf]
PhD Thesis, 2013.

[9] Badrinarayanan V., Budvytis, I., Cipolla, R.,
Semi-Supervised Video Segmentation using Tree Structured Graphical Models. [pdf]
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013.

[8] Badrinarayanan V., Budvytis I., Cipolla, R.,
Semi-Supervised Video Segmentation using Random Forests. [to buy]
Book Chapter in Decision Forests: for Computer Vision and Medical Image Analysis, Edited by Criminisi A., Shotton J., Springer 2013.

[7] Budvytis, I., Badrinarayanan, V., Cipolla, R.,
MoT - Mixture of Trees Probabilistic Graphical Model for Video Segmentation. [pdf] [oral presentation]
In Proc. British Machine Vision Conference, Surrey, September 2012

[6] Kim, T-K., Budvytis, I., Cipolla, R.,
Making a Shallow Network Deep: Conversion of a Boosting Classifier into a Decision Tree by Boolean Optimisation. [pdf]
International Journal of Computer Vision (IJCV), BMVC10 special issue, June 2011.

[5] Budvytis, I., Badrinarayanan, V., Cipolla, R.,
Semi-Supervised Video Segmentation using Tree Structured Graphical Models. [pdf]
In Proc. IEEE Conference on Computer Vision and Pattern Recognition, Colorado Springs, June 2011.

[4] Budvytis I., Badrinarayanan V., Cipolla R.,
Label Propagation in Complex Video Sequences using Semi-Supervised Learning. [pdf]
In Proc. British Machine Vision Conference, Aberystwyth, September 2010.

[3] Budvytis* I., Kim*, T-K., Cipolla R.,
Making a Shallow Network Deep: Growing a Tree from Decision Regions of a Boosting Classifier. [pdf] [oral presentation]
In Proc. British Machine Vision Conference, Aberystwyth, September 2010.
* indicates equal contribution

[2] Budvytis, I., Scott, J., Butler, A.,
Compass-Based Automatic Picture Taking using SenseCam. [pdf]
In Adj. Proc. of Pervasive 2008, Sydney, May 2008.

[1] Blackwell, A.F., Bailey, G., Budvytis, I., Chen, V., Church, L., Dubuc, L., Edge, D., Linnap, M., Naudziunas, V. and Warrington, H.
Tangible interaction in a mobile context. [pdf]
Workshop on Tangible user interfaces in context and theory in CHI 2007, San Jose, California, May 2007.

Contact Me