Jacopo Cavazza

Post Doc
Postdoctoral Researcher

Research Lines

Center

Contacts

jacopo.cavazza@iit.it
+39 010 2897 431
+39 346 7664 779

About

I am a Post Doc at the Pattern Analysis and Computer Vision (PAVIS) department. 

I graduated cum laude in 2014 from the Universita' degli Studi di Genova in Applied Mathematics.

Projects

Transfer Learning: Domain Adaptation & Zero-Shot Learning

Deep Learning

Kernel-based methods for Regression

Humans' Action Recognition and Intention Prediction

Crowd Counting

 Regularization methods

IIT Publications

  • 2021
  • Marmoreo F.iit, Cavazza J.iit, Murino V.iit

    Transductive Zero-Shot Learning by Decoupled Feature Generation

    Proceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021
  • 2020
  • Greco D., Cavazza J.iit, Del Bue A.

    Are Multiple Cross-Correlation Identities better than just Two? Improving the Estimate of Time Differences-of-Arrivals from Blind Audio Signals

    International Conference on Pattern Recogntion
  • Zunino A.iit, Cavazza J.iit, Volpi R.iit, Morerio P.iit, Cavallo A.iit, Becchio C.iit, Murino V.iit
    DOI

    Predicting Intentions from Motion: The Subject-Adversarial Adaptation Approach

    International Journal of Computer Vision, vol. 128, (no. 1), pp. 220-239
  • Paoletti G., Cavazza J.iit, Beyan C.iit, Del Bue A.iit

    Subspace Clustering for Action Recognition with Covariance Representations and Temporal Pruning

    International Conference on Pattern Recogntion
  • Bossi F., Willemse C., Cavazza J.iit, Marchesi S.iit, Murino V., Wykowska A.iit
    DOI

    The human brain reveals resting state activity patterns that are predictive of biases in attitudes toward robots

    Science Robotics
  • 2019
  • Bustreo M.iit, Cavazza J.iit, Murino V.iit

    Enhancing Visual Embeddings through Weakly Supervised Captioning for Zero-Shot Learning

    IEEE International Conference on Computer Vision Workshops (ICCVw), 1st International Workshop on Multi-Discipline Approach for Learning Concepts - Zero-Shot, One-Shot, Few-Shot and Beyond
  • Cavazza J.iit, Morerio P.iit, Murino V.iit
    DOI

    Scalable and compact 3D action recognition with approximated RBF kernel machines

    Pattern Recognition, vol. 93, pp. 25-35
  • 2018
  • Cavazza J.iit, Haeffele B.D., Lane C., Morerio P.iit, Murino V.iit, Vidal R.

    Dropout as a low-rank regularizer for matrix factorization

    International Conference on Artificial Intelligence and Statistics, AISTATS 2018, pp. 435-444
  • Morerio P.iit, Cavazza J.iit, Murino V.iit

    Minimal-entropy correlation alignment for unsupervised deep domain adaptation

    6th International Conference on Learning Representations, ICLR 2018 - Conference Track Proceedings
  • Roy A.iit, Cavazza J.iit, Murino V.iit

    Visually-Driven Semantic Augmentation for Zero-Shot Learning

    British Machine Vision Conference
  • 2017
  • Cavazza J.iit, Morerio P.iit, Murino V.iit
    DOI

    A compact kernel approximation for 3D action recognition

    Lecture Notes in Computer Science, vol. 10484 LNCS, pp. 211-222
  • Morerio P.iit, Cavazza J.iit, Volpi R.iit, Vidal R., Murino V.iit
    DOI

    Curriculum Dropout

    Proceedings of the IEEE International Conference on Computer Vision, vol. 2017-October, pp. 3564-3572
  • Zunino A.iit, Cavazza J.iit, Koul A.iit, Cavallo A.iit, Becchio C.iit, Murino V.iit
    DOI

    Predicting human intentions from motion cues only: A 2D+3D fusion approach

    MM 2017 - Proceedings of the 2017 ACM Multimedia Conference, pp. 591-599
  • Zunino A.iit, Cavazza J.iit, Murino V.iit
    DOI

    Revisiting human action recognition: Personalization VS. Generalization

    Lecture Notes in Computer Science, vol. 10484 LNCS, pp. 469-480
  • Zunino A.iit, Cavazza J.iit, Koul A.iit, Cavallo A.iit, Becchio C.iit, Murino V.iit
    DOI

    What Will i Do Next? the Intention from Motion Experiment

    IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, vol. 2017-July, pp. 1-8
  • Cavazza J.iit, Morerio P.iit, Murino V.iit
    DOI

    When Kernel Methods Meet Feature Learning: Log-Covariance Network for Action Recognition from Skeletal Data

    IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, vol. 2017-July, pp. 1251-1258
  • 2016
  • Cavazza J.iit, Zunino A.iit, Biagio M.S.iit, Murino V.iit
    DOI

    Kernelized covariance for action recognition

    Proceedings - International Conference on Pattern Recognition, pp. 408-413
  • 2015
  • Cavazza J.iit, Murino V.iit

    People Counting by Huber Loss Regression

    International Conference on Computer Vision Workshops