News

2026

• A Universal Encoder for Embodied Perception: Blog post and video for our recent work on universal robotic perception.
• New preprint online: Task Alignment: A simple and effective proxy for model merging in computer vision. [paper]
• Paper accepted at ICLR 2026: "ELViS: Efficient Visual Similarity from Local Descriptors that Generalizes Across Domains". [paper, code]
• Co-organizing the first ever AI For Peace Workshop at ICLR 2026. Our goal is to create a safe space for AI researchers to discuss hard truths about our research, build community and support any efforts towards the right direction. Join us if you are in Rio de Janeiro for ICLR.
• Serving as an Area Chair for CVPR 2026.

2025

• Paper accepted at CVPR 2025: "DUNE: Distilling a Universal Encoder from Heterogeneous 2D and 3D Teachers".
   [paper, project page, code, poster]
• Paper accepted at CVPR 2025: "LPOSS: Label Propagation Over Patches and Pixels for Open-vocabulary Semantic Segmentation".
    [paper, code, HF demo, poster]
• Invited talk on "Tales from the quest for generalization" at CVC-UAB. Barcelona, 28 January 2025.
• Serving as an Area Chair for ICCV 2025.

2024

• Paper accepted at ECCV 2024: "UNIC: Universal Classification Models via Multi-teacher Distillation". [paper, code]
• Keynote speaker at Netherlands Conference on Computer Vision (NCCV 2024). Den Bosch, The Netherlands, 29th of May 2024.
• Invited talk at the Synthetic Data for Computer Vision Workshop at CVPR 2024.
• Paper accepted at CVPR 2024: "Label Propagation for Zero-shot Classification with Vision-Language Models". [paper, code]
• Co-organizing the first ever African Computer Vision Summer School (ACVSS). The inaugural occurrence will be held in Kenya, in July 2024. Fully funded for African students, open to all.
• Invited talk at the University of Bristol on "Improving Generalization with Generative AI and Test-time training". Part of the MaVi Seminar Series of the Machine Learning and Computer Vision Group.
• Paper accepted at ICLR 2024: "Weatherproofing Retrieval for Localization with Generative AI and Geometric Consistency" [paper, project page]
• Serving as an Area Chair for ECCV 2024.

2023

• Paper accepted at NeurIPS 2023: "Test-time Training for Matching-based Video Object Segmentation" [paper, project page]
• Serving as an Action Editor for the Transactions on Machine Learning Research (TMLR) journal.
• Serving as an Area Chair for NeurIPS 2023 and CVPR 2024.
• Paper accepted at CVPR 2023: "Fake it till you make it: Learning transferable representations from synthetic ImageNet clones" [paper, project page]
• Paper accepted at ICLR 2023 (spotlight): "No Reason for No Supervision: Improved Generalization in Supervised Models" [paper, project page & pretrained models, code]
• Paper accepted at SCIA 2023 and the L3D-IVU-2023 Workshop at CVPR 2023 : "Rethinking matching-based few-shot action recognition" [paper, project page & pretrained models, code]

2022

• Find me on Mastodon at @skamalas@sigmoid.social
• Invited talk on "Granularity-aware Adaptation for Image Retrieval over Multiple Tasks". Instance-Level Recognition Workshop (ECCV 2022).
• Invited talk on "Improving generalization for classification and retrieval tasks". Visual Recognition Group at CTU in Prague.
• Paper accepted at TPAMI: "PoseBERT: A Generic Transformer Module for Temporal 3D Human Modeling" [paper, code]
• Paper accepted at ECCV 2022: "Granularity-aware Adaptation for Image Retrieval over Multiple Tasks" [paper]
• Paper accepted at TMLR: "TLDR: Twin Learning for dimensionality reduction" [paper, code]
• Amanda and I curated an illustrated version of "The summer day" by Mary Oliver, with watercolors created by the DALLE-Mini model
• Paper accepted at ICLR 2022: "Learning Super-Features for Image Retrieval" [paper, code, Hugging Face Spaces Demo]
• Serving as an Area Chair for CVPR 2023
• Co-organizing the Wiki-M3L 2022: Wikipedia and Multi-Modal & Multi-Lingual Research workshop at ICLR 2022.
• Invited talks on "Advances in Self-supervised Learning and measuring generalization" [slides]:
  • January 13th, Inria Rennes (Linkmedia Speaks Science lecture series, virtual)
  • January 18th, National Technical University of Athens (Research Challenges in Computer Science 2022, virtual) [video]
  • February 9th, University of Edinburgh (Advanced Vision Graduate Course, virtual)

2021

• Our camper van Alppy celebrates 10k kilometers with us in six months 😍
• Invited talks in Autumn 2021: Inria Grenoble (THOTH team, virtual), CTU in Prague (VRG group), Inria Paris (WILLOW team), ParisTech (Imagine team), University of Leicester (CSE Computing Seminar, virtual), KAUST (VisionCAIR group, virtual)
• Paper accepted at 3DV 2021: "Leveraging MoCap Data for Human Mesh Recovery" [paper, blogpost]
• Gave a two-part guest lecture on image representations at the Hanoi University of Science & Technology (HUST) (virtual) [slides]
• Paper accepted at ICCV 2021: "Concept Generalization in Visual Representation Learning" [paper, code, project page]
• Co-organizing the PAISS 2021 summer school
• Co-organizing the ActivityNet Entities Object Localization Challenge at CVPR 2021
• Paper accepted at CVPR 2021: "Probabilistic Embeddings for Cross-Modal Retrieval" [paper, code]
• Invited talk at the Robotics Institute at Carnegie Mellon University on Self-supervised Learning and generalization (VASC Seminar Series, virtual)

2020

• Serving as an Area Chair for ICCV 2021 and CVPR 2021
• Paper accepted at NeurIPS 2020: Hard Negative Mixing for Contrastive Learning [paper, project page, blog post, slides, 3-min video]
• Call for an open PhD position with Prof. Giorgos Tolias at CTU in Prague. [update: position has now been filled]
• Paper accepted at ECCV 2020: Learning to Generate Grounded Image Captions without Localization Supervision [paper, code]
• New chapter: I joined NAVER LABS Europe in Grenoble, France as a research scientist
• Co-organizing the CV for Agriculture (CV4A) workshop at ICLR 2020

2019

• Paper accepted at ICCV 2019: Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution [paper, code]
• Tutorial and invited talk at Data Science Africa 2019 in Accra, Ghana:
  • Tutorial on Image representations and fine-grained recognition [normal resolution slides (~6.2MB), lower resolution slides (~1.6MB)]
  • Talk on Learnings from the Computer Vision for Global Challenges (CV4GC) initiative [slides]
• Four papers accepted at CVPR 2019:
  • Grounded Video Description (oral) [paper, ActivityNet-Entities dataset, code]
  • Graph-Based Global Reasoning Networks [paper, code]
  • DMC-Net: Generating Discriminative Motion Cues for Fast Compressed Video Action Recognition [paper, code]
  • Less is More: Learning Highlight Detection from Video Duration [paper]
• Co-organizing the Computer vision for Global Challenges (CV4GC) workshop at CVPR 2019
• Paper accepted at PAMI: Focal Visual-Text Attention for Memex Question Answering [paper, MemexQA Dataset]
• Paper accepted at AAAI 2019: Large-scale Visual Relationship Detection [paper, code]

2018 and older

• Paper accepted at NeurIPS 2018: A^2-Nets: Double Attention Networks [paper, code]
• Paper accepted at ECCV 2018: Multi-Fiber Networks for Video Recognition [paper, code]
• Interactive visual art installation Do Silhouettes Dream? on display from July 26th till August 2nd 2017 at the ArtScience Museum in Singapore [ short paper, code, short interview]
• New chapter: In February 2017 I joined Facebook Research in Menlo Park, California as a research scientist
• Our work on Similarity Search at Flickr just launched [News coverage: The Verge, Engadget, Petapixel, Digital Trends and Venture Beat]
• Best paper award at the LSCVS 2016 workshop at NeurIPS 2016 for Tag Prediction in Flickr: A view from the darkroom [paper]
• Paper accepted at CHI 2017: Multimodal Classification of Moderated Online Pro-Eating Disorder Content [paper]

Selected Publications

Full list at my Google Scholar profile.

2026

2025

2024

2023

2022

2021

2020

2019

• Y. Chen, H. Fan, B. Xu, Z. Yan, Y. Kalantidis, M. Rohrbach, Z. Yan, Y. Shuicheng, J. Feng
  Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution
  ICCV 2019
  [Numerous Opensource Implementations]
• Y. Chen, M. Rohrbach, Z. Yan, Y. Shuicheng, J. Feng, Y. Kalantidis
  Graph-Based Global Reasoning Networks
  CVPR 2019
• L. Zhou, Y. Kalantidis, X. Chen, J. Corso, M. Rohrbach
  Grounded Video Description
  CVPR 2019 (oral)
  [ActivityNet-Entities dataset and code]
• B. Xiong, Y. Kalantidis, D. Ghadiyaram, K. Grauman
  Less is More: Learning Highlight Detection from Video Duration
  CVPR 2019
• Z. Shou, Z. Yan, Y. Kalantidis, L. Sevilla-Lara, M. Rohrbach, X. Lin, S.-F. Chang
  DMC-Net: Generating Discriminative Motion Cues for Fast Compressed Video Action Recognition
  CVPR 2019
• L. Jiang, L. Cao, Y.~Kalantidis, S. Farfade and A. Hauptmann
  Focal Visual-Text Attention for Memex Question Answering
  TPAMI 2019
• J. Zhang, Y. Kalantidis, M. Rohrbach, M. Paluri A. Elgammal, M. Elhoseiny
  Large-Scale Visual Relationship Understanding
  AAAI 2019
  [code]

2018

• Y. Chen, Y. Kalantidis, J. Li, Y. Shuicheng, J. Feng
  A^2-Nets: Double Attention Networks
  NeurIPS 2018
• Y. Chen, Y. Kalantidis, J. Li, Y. Shuicheng, J. Feng
  Multi-Fiber Networks for Video Recognition
  ECCV 2018
  [code]

2017

• R. Krishna, Y. Zhu, O. Groth, J. Johnson, K. Hata, J. Kravitz, S. Chen, Y. Kalantidis, L.-J. Li, D.A. Shamma, M. Bernstein and L. Fei-Fei
  Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
  IJCV 2017
• L. Jiang, Y. Kalantidis, L. Cao, S. Farfade, J. Tang and A. Hauptmann
  Delving Deep into Personal Photo and Video Search
  WSDM 2017
• S. Chancellor, Y. Kalantidis, J. A. Pater, M. De Choudhury and D. A. Shamma.
  Multimodal Classification of Moderated Online Pro-Eating Disorder Content
  CHI 2017

2016

• P. Garrigues, S. Farfade, H. Izadinia, Kofi Boakye and Y. Kalantidis
  Tag Prediction in Flickr: A view from the darkroom
  Large Scale Computer Vision systems Workshop at NeurIPS 2016 (Best paper award)
• Y. Kalantidis, C. Mellina and S. Osindero
  Cross-dimensional Weighting for Aggregated Deep Convolutional Features
  Web-scale Vision and Social Media (VSM) Workshop, ECCV 2016
  [code]
• Y. Kalantidis, L. Kennedy, H. Nguyen, C. Mellina and D.A. Shamma
  LOH and behold: Web-scale visual search, recommendation and clustering using Locally Optimized Hashing
  Web-scale Vision and Social Media (VSM) Workshop, ECCV 2016
• Y. Kalantidis, A. Farahat, L. Kennedy, R. Baeza-Yates and D.A. Shamma
  Visual Congruent Ads for Image Search
  ICPR 2016

2011 - 2015

• Y. Avrithis, Y. Kalantidis, E. Anagnostopoulos and I. Z. Emiris
  Web-scale image clustering revisited
  ICCV 2015 (oral)
• Y. Kalantidis and Y. Avrithis
  Locally Optimized Product Quantization for Approximate Nearest Neighbor Search
  CVPR 2014
  [code]
• G. Tolias, Y. Kalantidis, and Y. Avrithis
  Towards large-scale geometry indexing by feature selection
  CVIU 2014
• Y. Kalantidis, L. Kennedy and L.-J. Li
  Getting the Look: Clothing Recognition and Segmentation for Automatic Clothing Suggestions in Everyday Photos
  ICMR 2013
• Y. Avrithis and Y. Kalantidis
  Approximate gaussian mixtures for large scale vocabularies
  ECCV 2012
• G. Tolias, Y. Kalantidis, and Y. Avrithis. Symcity: Feature selection by symmetry for large scale image retrieval. In ACM Multimedia (Oral paper) (ACM MM 2012), 2012.
• Y. Avrithis, Y. Kalantidis, G. Tolias, and E. Spyrou. Retrieving landmark and non-landmark images from community photo collections. In ACM Multimedia (Oral paper) (ACM MM 2010), 2010.
• Y. Avrithis, G. Tolias, and Y. Kalantidis. Feature map hashing: Sub-linear indexing of appearance and global geometry. In ACM Multimedia (Oral paper) (ACM MM 2010), 2010.
• Y. Kalantidis, LG. Pueyo, M. Trevisiol, R. van Zwol, and Y. Avrithis. Scalable triangulation-based logo recognition. In International Conference on Multimedia Retrieval (ICMR), 2011.
• Y. Kalantidis, G. Tolias, Y. Avrithis, M. Phinikettos, E. Spyrou, P. Mylonas, and S. Kollias. Viral: Visual image retrieval and localization. Multimedia Tools and Applications (MTAP), 2011.

Photography

Primarily shoot 35mm film and upload selected works on Flickr and recently also on Instagram.
Note: All rights for the photos below are reserved. Feel free to contact me for non-commercial, research or artistic uses, happy to provide them for such usecases.

Film

Have shot two films, an award winning short and a feature length documentary screened at the Athens International Film Festival (Νύχτες Πρεμιέρας) in 2015. Both are available online.

• Love is Blind (2008)

  Short film (8', black and white)

  

  Official Selection:

  • 11th International Panorama of Independent Film and Video Creators (Patra, Greece 2009)
  • Athens Fantastic Film Festival 2009 (Ilioupoli, Athens, Greece 2009)

  Awards:

  • Audience Choice Award (Athens Fantastic Film Festival 2009, Ilioupoli, Greece)
  • Online Fantastic Film Audience Choice Award (2010, www.bigbang.gr)

• Blind Crossing (2015)

  Documentary Feature Film (57', color)

  

  Visit the IMDB page.

  Official Selection:

  • Athens International Film Festival (Athens, Greece, 2015)
  • Chalkida DocFest (Chalikda, Greece, 2015)

Interactive art

Do Silhouettes Dream?

...was an interactive art installation, co-created with Clayton Mellina and Cheng Xu.

It was on display from July 26th till August 2nd 2017 at the ArtScience Museum in Singapore.

"Do Silhouettes Dream?" aims to place fascinating results of cutting-edge artificial intelligence in a public embodied experience. As participants stand in front of a projection screen, their silhouettes create an area of hallucination that changes with their movement. Hallucinations may vary; some are ``dreams'', transporting the area covered by the silhouette to past or future times. Others escape reality and either alter the visual artistic style or form ``deep dreams'', i.e. hallucinations created through a feedback loop on artificial deep neural networks.

More information can be found in this short paper, while code to reproduce the installation is available on Github.

The artwork was accepted in the Art Track of the Creativity and Cognition 2017 conference as part of the Microbites of Innovation showcase.

You may watch a short interview with Cheng and me here.

Some photos and videos here!

(*) Figures above feature a Creative Commons image courtesy of Mike Cartmell on Flickr and a frame from Singapore City Day-Night Time lapse, used with permission.

Music

Ex Keyboard/piano player, lover of improvisational music (aka always lazy to study). Have played with many bands over the years [citation needed] spanning different genres, e.g. Greek rock with Amorfi Plektani, pop-rock with λaternative, jazz with the Silly Walks, and improvisational folk-rock with Dimitris Laios. Fun fact: Played with the House Band at the CVPR 2023 reception. It was a lot of fun and definitely one of the largest gigs ever. So, have now technically giged in New Orleans :) Here is a tweet with some basic proof.

Disturbing Music: In 2014, curated a web-radio series dedicated to experimental and improvisational music. All playlist can be found in the Disturbing Music Blog.

Traveling

Lover of road trips, occasional car/van dweller (see also Alppy the Van below). Below is a map of countries visited with red dots denoting current and past homes.

Surfing

Hopelessly in love with surfing, despite being eternally a beginner. The map below shows some of the surf spots surfed over the years.

Alppy the Van

In March 2021, and after years of planning, we adopted a 2016 Opel Vivaro that was living quietly under the shadow of the Alps. During the Summer of 2021 and with the help of Dionysis the woodsmith, Alppy the Van (they/them) was transformed from a 9-seater van to a home. An old canine friend, Alppy Junior, is excited to be our loyal companion and up for any adventure!

Alppy News
> Mar 2021: Alppy the Van finds a new household
> Jul 2021: The first version of Alppy as a home is completed 😍
> Aug 2021: First roadtrip with Alppy around the Peloponnese, the West coast of Greece and Toscany, Italy
> Sept 2021: In less than 6 months, Alppy celebrates 10k kilometers with us 💗! Cheers to many, many more!