HAUTVisual servoing

The various points studied in this theme are listed below as links leading to sections with their details and illustrations:

Direct visual servoing | Camera control

HAUTDirect visual servoing

  • Visual Path planning :
    A visual path is defined by a sequence of images, indirectly describing the poses a robot much go through. Successive direct visual servoings allow to follow the visual path. Instead of having a human pilot teaching to the robot the path to follow, this work automatically plans the ideal visual path, i.e. the set of images from a start pose to a goal pose, maximizing its ability to be followed:
    Drone and visual path

    International Journal

    E. A. Rodriguez Martinez, G. Caron, C. Pégard, D. Lara Alabazares, Photometric-Planner for Visual Path Following, IEEE Sensors Journal, September 2020. PDF | Video

    International conference

    E. A. Rodriguez Martinez, G. Caron, C. Pégard, D. Lara Alabazares, Photometric Path Planning for Vision-based Navigation, IEEE International Conference on Robotics and Automation, ICRA'20, Accepted, to appear, Paris, France, May 2020. PDF | Video

    French communication

    E. A. Rodriguez Martinez, G. Caron, C. Pégard, D. Lara Alabazares, Photometric Path Planning for Vision-based Navigation, Reconnaissance des Formes, Image, Apprentissage et Perception, RFIAP'20, Vannes (virtual), June 2020. | Video presentation

  • Photometric Gaussian Mixtures-based visual servoing:
    Exploiting the new Photometric Gaussian Mixture (PGM) image transform and, furthermore, adding the image smoothing amount as an additional degree of freedom, increases significantly the convergence domain of direct visual servoing:

    International Journal

    N. Crombez, G. Caron, E. Mouaddib, F. Chaumette, Visual Servoing with Photometric Gaussian Mixtures as Dense Features, IEEE Transactions on Robotics, 35(1):49-63, February 2019. PDF

    International conference

    N. Crombez, G. Caron, E. Mouaddib, Photometric Gaussian Mixtures-based Visual Servoing, IEEE/RSJ International Conference on Intelligent RObots and Systems, IROS'15, pp. 5486-5491, Hamburg, Germany, October 2015. PDF

    French conference

    N. Crombez, G. Caron, E. Mouaddib, Asservissement visuel basé mélanges de gaussiennes photométriques, Reconnaissance de Formes et Intelligence Artificielle, RFIA'16, Clermont-Ferrand, June 2016.

  • Omnidirectional photometric visual servoing:
    Visual servoing uses generally geometric features (points, lines, ...) and aims to move the camera minimizing the error between current detected features an desired ones. The reference image is acquired as an initial step at the desired position, which is a kind of supervised learning. The idea, here, is to directly use intensities of all pixels as "features". This work is the extension of previous works of Lagadic team. Here is a video where one can see the robot from an external viewpoint, images acquired by the fisheye camera, camera trajectory and difference between current and desired images:
    This video can also be downloaded, as the previous one (IROS 2010).

    International journal

    Guillaume Caron, Eric Marchand and El Mustapha Mouaddib, Photometric visual servoing for omnidirectional cameras, Autonomous Robots, 35(2):177-193, October 2013. PDF

    International Conferences

    Y. Alj, G. Caron, Featureless omnidirectional vision-based control of non-holonomic mobile robot, KROS-IEEE/RSJ 12th Int. Conf. on Ubiquitous Robots and Ambient Intelligence, URAI'15, Goyang, South Korea, October 2015. PDF

    Bertrand Delabarre, Guillaume Caron and Eric Marchand, Omnidirectional Visual Servoing using the Normalized Mutual Information, 10th IFAC Symposium on Robot Control, SYROCO'12, Dubrovnik, Croatia, September 2012. PDF

    Guillaume Caron, Eric Marchand and El Mustapha Mouaddib, Omnidirectional Photometric Visual Servoing, IEEE/RSJ International Conference on Intelligent RObots and Systems, IROS'10, Taipei, Taiwan, October 2010. PDF

    National Talks

    Guillaume Caron, Eric Marchand and El Mustapha Mouaddib, Asservissement visuel photométrique en vision omnidirectionnelle, journée Perception et commande référencées capteurs non-convetionnels (GDR Robotique, CNRS), Paris, France, April 2010.
    Guillaume Caron, El Mustapha Mouaddib and Eric Marchand Asservissement visuel en vision omnidirectionnelle, journée Rencontre des chercheurs en Image, Vision et Reconnaissance de Formes des régions Nord-Pas-De-Calais et Picardie (Telecom Lille 1), Lille, France, June 2010.

HAUTVisual relevance-based camera control

  • International conferences

    Z. Habibi, E. Mouaddib, G. Caron, Good feature for framing: saliency-based Gaussian mixture, IEEE/RSJ International Conference on Intelligent RObots and Systems, IROS'15, pp. 3682 - 3687, Hamburg, Germany, October 2015.

    Z. Habibi, G. Caron, E. Mouaddib, Assistive visual framing in 3D dense Points Cloud, IEEE Digital Heritage Int. Congress, pp. 109 - 112, Granada, Spain, October 2015. PDF

    International workshop

    Z. Habibi, G. Caron, E. Mouaddib, 3D model automatic exploration: Smooth and Intelligent Virtual Camera Control, ACCV 2014 workshop on e-Heritage, Singapore, November 2014. PDF

    French conferences

    Z. Habibi, E. Mouaddib, G. Caron, Mélange de gaussiennes basées saillance pour le cadrage visuel, Reconnaissance de Formes et Intelligence Artificielle, RFIA'16, Clermont-Ferrand, June 2016.

    Z. Habibi, G. Caron, E. Mouaddib, Exploration réaliste et pertinente d'un nuage de points 3D dense et coloré, Congrès des jeunes chercheurs en vision par ordinateur, ORASIS'15, Amiens, June 2015. PDF

    French communication

    Z. Habibi, G. Caron, E. Mouaddib, Exploration automatique d'un environnement 3D : contrôle fluide et intelligent de caméra virtuelle, Reims Image, Reims, November 2014. PDF