#cvpr2024 — Public Fediverse posts

First was the 4th workshop of adversarial machine learning on computer vision at #CVPR2024. I highly recommend the whole event, with standout talks by Zico Kolter (adversarial attacks on aligned language models, talk of the year candidate) and Ludwig Schmidt (data-centric view on reliable generalization) https://www.youtube.com/watch?v=U3SiUQvZ5LM (2/7) #AI #cybersecurity

First was the 2nd workshop on large models at #CVPR2024, with notable talks by Yue Fan (toward a diffusion-based generalist for dense vision tasks) and Hao Su (generative and understanding 3D objects in an open world) https://www.youtube.com/watch?v=UbsRsFkXs8M (2/4) #AI

First was the 4th workshop on open world vision at #CVPR2024. I highly recommend the whole event, and I particularly liked the talks by Walter Scheirer (open issues in open world learning) and Deva Ramanan (open world learning and large mutlimodal models) https://www.youtube.com/watch?v=eTBRrMn9fOU (2/9) #AI

Next was the 2nd workshop on generative models for computer vision at #CVPR2024, with notable talks by Katerina Fragkiadaki (image and video perception with generative feedback) and Andrea Vedaldi (3D generative AI) https://www.youtube.com/watch?v=GEaSNJOpfsU (5/9) #AI

Next was the 3rd explainable AI for computer vision workshop at #CVPR2024. I particularly liked the talks by Bernt Schiele (inherent interpretability for deep learning in computer vision) and Tim Miller (human-centered counterfactual explanations for image classification) https://www.youtube.com/watch?v=o2YmzPXtAgc (3/4) #AI #XAI

First were a pair of talks by Siwei Lyu (deepfake detection in the real world) and Venkatesh Babu Radhakrishnan (uncovering and addressing biases in diffusion models) at the #CVPR2024 workshop on fair, data-efficient, and trusted computer vision https://www.youtube.com/watch?v=1EZaWcqqlmo (2/4) #AI

Next was a great talk by Song Han on designing efficient visual language models at #CVPR2024 https://www.youtube.com/watch?v=lirjgIgngUE (3/5) #AI

First was the workshop on computer vision in the built environment at #CVPR2024. I particularly liked the talk by Caitlin Mueller on using a variety of innovative algorithmic methods to design more sustainable structures https://www.youtube.com/watch?v=kPX459hvku4 (2/6) #AI #architecture

First was an incredible pair of talks at #CVPR2024 by @sarameghanbeery (benchmarking models in a changing world) and William Agnew (mapping the computer vision surveillance and weapons pipeline). These are both tremendous talks, interrogating the fundamental issues with classic computer science benchmarking orthodoxy and the increasing adoption of "neutral" computer vision technologies in decidedly non-neutral contexts, respectively. Highly recommend https://www.youtube.com/watch?v=nPQ1pxOgurQ (2/3) #AI

First was the workshop on continual learning in computer vision at #CVPR2024, with notable talks by Vineeth Balasubramanian (integrating explainability and privacy-awareness) and Elisa Ricci (class-incremental novel class discovery) https://www.youtube.com/watch?v=fQ_HZu-8eMI (2/4) #AI

Next was a fantastic slate of talks at the #CVPR2024 workshop on responsible generative AI. I particularly liked the talks by Björn Ommer (dimensions of responsible generative AI) and Miranda Bogen (intersection of organizational processes/goals and building responsible tech applications). Highly recommend https://www.youtube.com/watch?v=gJobLgvgabs (4/7) #AI #ethics

Last was an incredible pair of talks by Ludwig Schmidt (evaluating training sets for multimodal models) and Ranjay Krishna (evaluation of vision-language models) at #CVPR2024. I highly recommend both talks, and I particularly liked how Krishna shows the significant limitations of many popular benchmarks and just how poor large vision models are at scene composition https://www.youtube.com/watch?v=QX92Eg6nJEo (4/4) #AI

This morning at 9am pacific we welcome Chaehyeon Song, one of the authors of the CVPR paper "Unbiased Estimator for Distorted Conics in Camera Calibration." Chaehyeon will walk us through the paper and show us a live demo on the OpenCV Live! stream.

Watch along on YouTube: https://youtube.com/live/MTMMoN6ogcY
or Twitch: https://twitch.tv/opencv_

#OpenCV #ComputerVision #AI #MachineLearning #CVPR2024

Alibaba Cloud launches English version of AI model hub https://www.artificialintelligence-news.com/2024/06/25/alibaba-cloud-launches-english-version-ai-model-hub/ #alibaba #modelscope #cvpr2024 #ai #tech #news #technology

Alibaba Cloud launches English version of AI model hub https://www.artificialintelligence-news.com/2024/06/25/alibaba-cloud-launches-english-version-ai-model-hub/ #alibaba #modelscope #cvpr2024 #ai #tech #news #technology

Alibaba Cloud launches English version of AI model hub https://www.artificialintelligence-news.com/2024/06/25/alibaba-cloud-launches-english-version-ai-model-hub/ #alibaba #modelscope #cvpr2024 #ai #tech #news #technology

Alibaba Cloud launches English version of AI model hub https://www.artificialintelligence-news.com/2024/06/25/alibaba-cloud-launches-english-version-ai-model-hub/ #alibaba #modelscope #cvpr2024 #ai #tech #news #technology

Alibaba Cloud launches English version of AI model hub https://www.artificialintelligence-news.com/2024/06/25/alibaba-cloud-launches-english-version-ai-model-hub/ #alibaba #modelscope #cvpr2024 #ai #tech #news #technology

I’m happy to share our recent work on Creating a Digital Twin of Spinal Surgery: A Proof of Concept. We showcase how complete surgeries can be digitized in order to obtain a highly accurate and detailed spatio-temporal representation. We strongly believe that these surgical digital twins have significant potential in the areas of education and training, training surgical robots in simulation, synthetic data generation, and many more.

Project page: https://jonashein.github.io/surgerydigitization/

#CVPR2024
#digitalTwin

If you are following #CVPR2024 or even attending, check out what the Matthias Niessner TUM lab is presenting. Some really groundbreaking research and innovative work in computer vision, Gen AI and 3D!

Great paper compilation here: https://www.youtube.com/watch?v=POH-CW1uyhg

if you're at #cvpr2024, come listen to me pontificate about generative AI and art today -- panel starts at 4pm, my talk at 4:40pm in Summit 343

https://cveu.github.io

This morning we welcome back multi-time guests Paula Ramos and Samet Akcay to tell us what the OpenVINO team have in store for The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2024, which is being hosted in Seattle this year on June 17-21. OpenCV will be at CVPR as well, with fresh merch and demos in the exhibition hall.

Watch on YouTube, live: https://youtube.com/live/vwUB64GbUNA

#OpenCV #ComputerVision #AI #CVPR2024 #OpenSource

Thanks to the SCoFT team for all the great work that got our paper into #CVPR !! 🎉🎊

Authors: Zhixuan Liu, Peter Schaldenbrand, Beverley-Claire Okogwu, Wenxuan Peng, Youngsik Yun, Andrew Hundt, Jihie Kim, Jean Oh.

Our paper is:
SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation

The ArXiV paper link is below with all the details, and we expect to make code available soon!

Thanks for reading!

#CVPR #CVPR2024 #AI #ML #SCoFT

https://arxiv.org/abs/2401.08053

19/n

SCoFT has been accepted at #CVPR! 🎉🎊

Remember Google Gemini’s biased medieval England generated images that were just everywhere? Ancient internet history, I know.

I've been chomping at the bit bc we've had methods for more culturally sensitive image generation under review that help to address the kinds of issues we saw with Gemini!

SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation

#CVPR2024

https://arxiv.org/abs/2401.08053

🧵 1/n

Where are my goats?! See you @CVPR with an exciting program and hope to see many animal enthusiasts in Seattle #CVPR2024! https://cv4animals.com #animal #behaviouralscience #computervision #CV4Animals

Where are my goats?! See you @CVPR with an exciting program and hope to see many animal enthusiasts in Seattle #CVPR2024! https://cv4animals.com #animal #behaviouralscience #computervision #CV4Animals

Where are my goats?! See you @CVPR with an exciting program and hope to see many animal enthusiasts in Seattle #CVPR2024! https://cv4animals.com #animal #behaviouralscience #computervision #CV4Animals

Where are my goats?! See you @CVPR with an exciting program and hope to see many animal enthusiasts in Seattle #CVPR2024! https://cv4animals.com #animal #behaviouralscience #computervision #CV4Animals

Where are my goats?! See you @CVPR with an exciting program and hope to see many animal enthusiasts in Seattle #CVPR2024! https://cv4animals.com #animal #behaviouralscience #computervision #CV4Animals

I am happy to announce that WE HAVE ONE MORE WORKSHOP at
CVPR

"Urban Scene Modeling: Where Vision Meets Photogrammetry and Graphics"
https://usm3d.github.io

Or "what useful thing you can do after SfM"
P.S. We will have 2(!) challenges with money, stay tuned.
#CVPR2024