News
04/09/2017 One paper accepted at TPAMI
03/05/2017 I have
successfully defended my thesis on Continuous Regression.
01/11/2016 Continuous
Regression site launched: Continuous
Regression site
iCCR
code
NOTE: To use the Matlab Test on Recorded Videos ,
you have to load into the workspace the model.mat and
the example_params.mat, then call
> load model.mat
> load example_params.mat
> video = [];
> data = track( model, video, params )
- params is included in the example_params.mat,
and video is the path to the video to test. If video = [], then the tracker will
get the input stream from the webcam. Data will contain all the tracked points.
Should you find any problem running the code, please do not hesitate in contact
me at: psxes1[removethis]_at_
.nottingham.ac.uk Updated: Linux bug has been found and fixed.
iCCR
Results on 300VW
HERE you can directly
download the tracked points for the 300VW Dataset, for each of the categories.
You will find a data_catX_iCCR_CCR.mat
file, where X will be A, B and C, corresponding to categories 1,2 and 3, respectively. Each is a struct
containing the following fields:
- data_iccrX{i}.tracked_pts : a 66 x 2 x NumberOfFrames
- data_iccrX{i}.video : the
corresponding video id
- data_iccrX{i}.nc : the indices of the frames that were removed from
evaluation, according to the 300VW organisers
- data_iccrX{i}.nframes : the total number of frames of the corresponding
video
Publications
- A Functional
Regression approach to Facial Landmark Tracking
- Sánchez-Lozano, E., Tzimiropoulos,
G., Martinez, B., De la Torre, F., Valstar,
M.
- IEEE Trans. on Pattern Analysis and Machine
Intelligence. 2017
- FERA 2017 – Addressing
Head Pose in the Third Facial Expression Recognition and Analysis
Challenge
- Valstar, M., Sánchez-Lozano, E., Cohn, J., Jeni, L., Girard, J., Zhang, Z., Yin, L., Pantic,
M.
- 12th IEEE Int’l Conference on Automatic Face &
Gesture Recognition (FG’17)
- Cascaded
Continuous Regression for Real-time Incremental Face Tracking [bib][code][code(demo)]
- Sánchez-Lozano, E., Martinez,
B., Tzimiropoulos, G., Valstar,
M.
- 14th European Conference on Computer Vision (ECCV'16).
(26% acceptance rate)
- Cascaded
Regression with Sparsified Feature Covariance
Matrix for Facial Landmark Detection [bib][code]
- Sánchez-Lozano, E., Martinez,
B., Valstar, M.F.
- Pattern Recognition Letters. Vol 73, pp. 19-25. 2016
- Audiovisual
Three-Level Fusion for Continuous Estimation of Russell's Emotion Circumplex [bib]
- Sánchez-Lozano, E., Lopez-Otero,
P., Docio-Fernandez, L., Argones-Rúa, E.,
Alba-Castro, J.L.
- 3rd Audio-Visual Depression and Emotion Recognition
Challenge (AVEC'13)
- Blockwise
Linear Regression for Face Alignment [bib]
- Sánchez-Lozano, E., Argones-Rúa, E., Alba-Castro,
J.L.
- 24th British Machine Vision Conference (BMVC'13) (30%
acceptance rate)
- Continuous
Regression for Non-Rigid Image Alignment [bib]
- Sánchez-Lozano, E., De la Torre, F.,
González-Jiménez, D.
- 12th European Conference on Computer Vision (ECCV'12)
(28% acceptance rate)
Parameter Constraints for Accurate Face Alignment in the Simultaneous Inverse Composition al Algorithm [bib]- Sánchez-Lozano, E., González-Jiménez, D.,
Alba-Castro, J.L.,
- 7th Int'l Sy mposium on Image and Signal Processing and Analysis
(ISPA'11) (56% acceptance rate)