home.social

#cvpr2023 — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #cvpr2023, aggregated by home.social.

  1. During #CVPR2023 @MetaAI
    presented Project Aria, a project that utilizes a sensor packed headset to help researchers gather data from the user’s perspective. As a part of their tutorials they included a presentation on how to use Rerun for visualizing multi-modal data! 🤯

  2. What is the latest research in quantifying animal movement and behavior?

    I wrote down a detailed overview here, based on what I saw at CVPR this year:
    writings.lambdaloop.com/posts/

    #PoseEstimation #animals #computervision #cvpr #cvpr2023 #neuroscience

  3. Enjoy Image Matching Challenge 2023 recap:

    ducha-aiki.github.io/wide-base

    tl;dr:
    - SfM is not solved
    - global desc similarity is hard
    - orientation invariance for SuperGlue is easy
    - PixSfM is good idea, but need follow-ups
    - KeyNetAffNet
    @kornia_foss rocks

    #IMC2023 #CVPR2023

    @cvpr

    #computervision #deeplearning
    #dmytrotweetsaboutDL

  4. Enjoy Image Matching Challenge 2023 recap:

    ducha-aiki.github.io/wide-base

    tl;dr:
    - SfM is not solved
    - global desc similarity is hard
    - orientation invariance for SuperGlue is easy
    - PixSfM is good idea, but need follow-ups
    - KeyNetAffNet
    @kornia_foss rocks

    #IMC2023 #CVPR2023

    @cvpr

    #computervision #deeplearning
    #dmytrotweetsaboutDL

  5. Enjoy Image Matching Challenge 2023 recap:

    ducha-aiki.github.io/wide-base

    tl;dr:
    - SfM is not solved
    - global desc similarity is hard
    - orientation invariance for SuperGlue is easy
    - PixSfM is good idea, but need follow-ups
    - KeyNetAffNet
    @kornia_foss rocks

    #IMC2023 #CVPR2023

    @cvpr

    #computervision #deeplearning
    #dmytrotweetsaboutDL

  6. Enjoy Image Matching Challenge 2023 recap:

    ducha-aiki.github.io/wide-base

    tl;dr:
    - SfM is not solved
    - global desc similarity is hard
    - orientation invariance for SuperGlue is easy
    - PixSfM is good idea, but need follow-ups
    - KeyNetAffNet
    @kornia_foss rocks

    #IMC2023 #CVPR2023

    @cvpr

    #computervision #deeplearning
    #dmytrotweetsaboutDL

  7. Enjoy Image Matching Challenge 2023 recap:

    ducha-aiki.github.io/wide-base

    tl;dr:
    - SfM is not solved
    - global desc similarity is hard
    - orientation invariance for SuperGlue is easy
    - PixSfM is good idea, but need follow-ups
    - KeyNetAffNet
    @kornia_foss rocks

    #IMC2023 #CVPR2023

    @cvpr

    #computervision #deeplearning
    #dmytrotweetsaboutDL

  8. HierVL is a new #AI approach injecting semantics into visual representations by capturing long-term and short-term associations in video-language embeddings capturing both 'what' and 'why' of human actions. 🤯

    vision.cs.utexas.edu/projects/

    #ComputerVision #DeepLearning #CVPR2023

  9. PanoHead: a breakthrough in 3D human head synthesis

    This novel generative #AI model produces 360° view-consistent images of full heads and overcomes limitations in existing methods by leveraging 3D GANs and innovative image alignment techniques. 🤩

    sizhean.github.io/panohead

    #CVPR2023 #DeepLearning #GenAI

  10. Rodney Brooks #CVPR2023 keynote about his PhD in the 70s: “In four years I processed three images.”

  11. made it to vancouver for #cvpr2023, really excited to host our 5th annual VizWiz Workshop, with three outstanding keynote speakers -- representing users, researchers, and industry --

    more here:
    vizwiz.org/workshops/2023-work

  12. Happy to announce that our paper "Inferring the Past: A Combined CNN-LSTM Deep Learning Framework To Fuse Satellites for Historical Inundation Mapping" was accepted at CVPR workshop Earthvision!

    openaccess.thecvf.com/content/

    #earthvision #CVPR2023

  13. We released our Neural Parametric Head Models (NPHM) dataset from our #CVPR2023 paper!

    It includes over 5600 high-fidelity 3D scans of human heads from 272 subjects - all publicly available!

    Check it out!
    simongiebenhain.github.io/NPHM

    We've also released the NPHM code: github.com/SimonGiebenhain/NPH

    We provide several examples and a small demo how to use our dataset. It's now super easy to build your own parametric face/head model!

    Great work by Simon, Tobias, Markos Martin Lourdes!

  14. D2Former: Jointly Learning Hierarchical Detectors and Contextual Descriptors via Agent-Based Transformers

    Jianfeng He, Yuan Gao, Tianzhu Zhang, Zhe Zhang, Feng Wu

    tl;dr: no idea how that works, hierarchical attention something. No eval on #IMC

    openaccess.thecvf.com/content/
    #CVPR2023
    #computervision #deeplearning
    #dmytrotweetsaboutDL

  15. D2Former: Jointly Learning Hierarchical Detectors and Contextual Descriptors via Agent-Based Transformers

    Jianfeng He, Yuan Gao, Tianzhu Zhang, Zhe Zhang, Feng Wu

    tl;dr: no idea how that works, hierarchical attention something. No eval on #IMC

    openaccess.thecvf.com/content/
    #CVPR2023
    #computervision #deeplearning
    #dmytrotweetsaboutDL

  16. D2Former: Jointly Learning Hierarchical Detectors and Contextual Descriptors via Agent-Based Transformers

    Jianfeng He, Yuan Gao, Tianzhu Zhang, Zhe Zhang, Feng Wu

    tl;dr: no idea how that works, hierarchical attention something. No eval on #IMC

    openaccess.thecvf.com/content/
    #CVPR2023
    #computervision #deeplearning
    #dmytrotweetsaboutDL

  17. D2Former: Jointly Learning Hierarchical Detectors and Contextual Descriptors via Agent-Based Transformers

    Jianfeng He, Yuan Gao, Tianzhu Zhang, Zhe Zhang, Feng Wu

    tl;dr: no idea how that works, hierarchical attention something. No eval on #IMC

    openaccess.thecvf.com/content/
    #CVPR2023
    #computervision #deeplearning
    #dmytrotweetsaboutDL

  18. D2Former: Jointly Learning Hierarchical Detectors and Contextual Descriptors via Agent-Based Transformers

    Jianfeng He, Yuan Gao, Tianzhu Zhang, Zhe Zhang, Feng Wu

    tl;dr: no idea how that works, hierarchical attention something. No eval on #IMC

    openaccess.thecvf.com/content/
    #CVPR2023
    #computervision #deeplearning
    #dmytrotweetsaboutDL

  19. SFD2: Semantic-guided Feature Detection and Description

    Fei Xue, Ignas Budvytis, Roberto Cipolla

    tl;dr: more principled approach to use segmentation for image matching, than class-name filtering.

    arxiv.org/abs/2304.14845.pdf

    #computervision #deeplearning
    #dmytrotweetsaboutDL
    #CVPR2023

  20. Check out "Learning Neural Parametric Head Models"!

    One of the best parts of writing a paper is the collaboration with great talents. At Synthesia we are working with TUM+UCL and our first publication will be presented at #CVPR2023 🎉 Special thanks to the first author Simon (simongiebenhain.github.io) for the fantastic execution.

    Project page: simongiebenhain.github.io/NPHM

  21. Excited to share "Learning Neural Parametric Head Models" #CVPR2023!

    We capture over 5200 high-quality 3D human head scans from which we build a neural parametric head model that disentangles & expressions and deformations.

    Core of our representation is an ensemble of local MLPs to facilitate high geometric detail. We will release the entire data asap!

    simongiebenhain.github.io/NPHM
    youtu.be/0mDk2tFOJCg

  22. Camera frame registration with little or no overlap?

    Check out Can's ObjectMatch: We leverage object semantics to find indirect correspondences between frames. Great for any SLAM and SfM pose optimization!

    Video: youtu.be/kuXoKVrzURk
    Project: cangumeli.github.io/ObjectMatc

    #CVPR2023

  23. Excited to share Norman's DiffRF: Rendering-guided 3D Radiance Field Diffusion #CVPR2023 highlight!

    2D diffusion is great, but what about 3D? We show radiance field diffusion with rendering guidance for consistent and editable 3D synthesis.

    Video: youtu.be/qETBcLu8SUk
    Project: sirwyver.github.io/DiffRF

    Great collaboration with Peter's group at Meta!

  24. PMatch: Paired Masked Image Modeling for Dense Geometric Matching

    Shengjie Zhu, Xiaoming Liu

    tl;dr: cross-view patch reconstruction pre training, then use (modified) LoFTR for two view matching.
    Code repo is not yet created.
    arxiv.org/abs/2303.17342.pdf
    #computervision #deeplearning
    #dmytrotweetsaboutDL #CVPR2023

  25. 3D Line Mapping Revisited

    Shaohui Liu, Yifan Yu, Rémi Pautrat, Marc Pollefeys, Viktor Larsson

    tl;dr: Point-guided line SfM with scale-invariant scoring -> profit. Even improves SuperPoint + Colmap bundle adjustment

    arxiv.org/abs/2303.17504.pdf

    #CVPR2023

    #computervision #deeplearning
    #dmytrotweetsaboutDL #CVPR2023

  26. Working on an interesting fine-grained problem, but the #ICCV2023 deadline is approaching too quickly?
    The FGVC10 paper submission website is now open!

    sites.google.com/view/fgvc10/s

    Deadline: March 20

    #CVPR #CVPR2023 #FGVC

  27. Working on an interesting fine-grained problem, but the #ICCV2023 deadline is approaching too quickly?
    The FGVC10 paper submission website is now open!

    sites.google.com/view/fgvc10/s

    Deadline: March 20

    #CVPR #CVPR2023 #FGVC

  28. Working on an interesting fine-grained problem, but the #ICCV2023 deadline is approaching too quickly?
    The FGVC10 paper submission website is now open!

    sites.google.com/view/fgvc10/s

    Deadline: March 20

    #CVPR #CVPR2023 #FGVC

  29. Working on an interesting fine-grained problem, but the #ICCV2023 deadline is approaching too quickly?
    The FGVC10 paper submission website is now open!

    sites.google.com/view/fgvc10/s

    Deadline: March 20

    #CVPR #CVPR2023 #FGVC

  30. Working on an interesting fine-grained problem, but the #ICCV2023 deadline is approaching too quickly?
    The FGVC10 paper submission website is now open!

    sites.google.com/view/fgvc10/s

    Deadline: March 20

    #CVPR #CVPR2023 #FGVC

  31. NICE! High-Res Facial Appearance Capture from Polarized Smartphone Images 🤳

    This approach uses a smartphone with a cheap polarization foil to capture facial textures in dark settings to then create photo-realistic 3D #DigitalHumans 👉dazinovic.github.io/polface/

    #AI #CVPR23 #CVPR2023 #Computervision #DeepLearning

  32. Great News! FGVC10 (10th Workshop on Fine-Grained Visual Categorization) was accepted to CVPR 2023 and will take place in Vancouver. Interested in fine-grained learning and its applications? Consider submitting your paper to FGVC10.

    Deadline: March 20.

    More information on the website: sites.google.com/view/fgvc10

    #CVPR #CVPR2023 #FGVC

  33. Congrats to everyone who has their workshop accepted to #CVPR2023 :)

    I regret to announce that #IMW2023 #IMC2023 is also accepted to CVPR so I will annoy the next half of the years with news about Image Matching Challenge 2023. Stay tuned :)