KAUST VISION CAIR (Computer Vision, Core AI Research) group led by Prof. @moElhoseiny. @KAUST_News https://xitter.cf/t.co/wpPoM4iiYj…
Thuwal, Saudi Arabia
Joined August 2020
- Tweets128
- Following72
- Followers342
- Likes116
KAUST Vision CAIR group retweeted
Prof @haosu_twitr sharing insights on training free generation using 3d diffusion. C3DV workshop at CVPR.
📣 Call for Papers: As part of C3DV, we invite researchers to submit papers on topics related to 3D compositional vision.
📅 Non-archival papers deadline: May 15th. bit.ly/4c2BsQD
🧵 2/4
KAUST Vision CAIR group retweeted
Prof. Angel Chang of SFU sharing her journey on compositional scene generation at C3DV workshop @CVPR
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2025! @CVPR
🏆 Challenges: bit.ly/4c8DfDS
📢 Call for Papers: bit.ly/4c2BsQD
C3DV will feature two exciting challenges and a fantastic lineup of speakers!
🧵👇
KAUST Vision CAIR group retweeted
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2025! @CVPR
🏆 Challenges: bit.ly/4c8DfDS
📢 Call for Papers: bit.ly/4c2BsQD
C3DV will feature two exciting challenges and a fantastic lineup of speakers!
🧵👇
KAUST Vision CAIR group retweeted
The C3DV @CVPR is starting at 8:15. Please join us,
Where: Room 110 B, Music City Center, Nashville
Zoom link is accessible through CVPR for registered participants, excellent lineup of speakers!
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2025! @CVPR
🏆 Challenges: bit.ly/4c8DfDS
📢 Call for Papers: bit.ly/4c2BsQD
C3DV will feature two exciting challenges and a fantastic lineup of speakers!
🧵👇
KAUST Vision CAIR group retweeted
200+ students from across Saudi Arabia explored #AI through art, music & design at the SAAI Factory Hackathon for Kids at #KAUST.
Their projects tackled inclusion, climate change & immersive learning—blending creativity with tech.
Learn more: cemse.kaust.edu.sa/articles/…
KAUST Vision CAIR group retweeted
#ICLR 2025
🚀 Excited to share that three papers have been accepted at ICLR 2025! 🎉 Huge thanks to my incredibly talented students and collaborators for their dedication and hard work—this wouldn't have been possible without you!
KAUST Vision CAIR group retweeted
@ICLR25: @KAUSTVisionCAIR’s wenxuan zhang and @ben_nebulous are presenting BFPO and Toddler diffusion this morning; posters 277 and 280 Hall 2; stop by to learn more about carefully modeling the dichotomy of safety and helpfulness and more interpretable and efficient diffusion.
#ICLR 2025
🚀 Excited to share that three papers have been accepted at ICLR 2025! 🎉 Huge thanks to my incredibly talented students and collaborators for their dedication and hard work—this wouldn't have been possible without you!
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2025! @CVPR
🏆 Challenges: bit.ly/4c8DfDS
📢 Call for Papers: bit.ly/4c2BsQD
C3DV will feature two exciting challenges and a fantastic lineup of speakers!
🧵👇
🏆 We are also hosting two 3D vision challenges this year! This year's highlight: the 3DCoMPaT-200 Challenge and the Language-Based Part Grounding Challenge, focusing on 3D object composition and understanding.
📥 C3DV Challenges: bit.ly/4c8DfDS
🧵 3/4
👥 This workshop is organized by: @habib__slim, @Meh_Mo0d, Abdulwahab Felemban, Wolfgang Heidrich, @vaphab, @natalianeverova, Junjie Fei, Xiang Li, @peter_wonka, and @moelhoseiny
#CVPR2025 #3DVision #CVPR @CVPR
🧵 4/4
KAUST Vision CAIR group retweeted
📌 Paper 3: Query-based Knowledge Transfer for Heterogeneous Learning Environments
@Norah Alballa, Wenxuan Zhang,@Ziquan Liu. Mohamed Elhoseiny. Marco Canini
✅ This paper introduces Query-based Knowledge Transfer (QKT), a novel framework for decentralized collaborative learning
KAUST Vision CAIR group retweeted
📌 Paper 2: Bi-Factorial Preference Optimization (BFPO): Balancing Safety-Helpfulness in Language Models
by Wenxuan Zhang, Philip Torr, Mohamed Elhoseiny*, Adel Bibi*,
✅ BFPO is a supervised learning framework that reformulates RLHF’s joint objective of safety and helpfulness..
KAUST Vision CAIR group retweeted
...into interpretable stages (e.g., contours, palettes, textures) and leverages Schrödinger Bridge optimal transport for seamless transitions between modalities, enhancing control and flexibility in the generation process.
KAUST Vision CAIR group retweeted
📌 Paper 1: ToddlerDiffusion: Interactive Structured Image Generation with Cascaded Schrödinger Bridge
by Eslam Bakr, Liangbing Zhao, Tao HU, Matthieu Cord, Patrick Pérez, Mohamed Elhoseiny
✅This paper introduces a new diffusion framework that decomposes RGB image generation
KAUST Vision CAIR group retweeted
Excellent collaboration with an amazing team at Meta. Huge thanks to @YoungXiong1 for hosting @xiaoqian_shen's internship!
@liuzhuang1234,@Hu_Hsu
@garvinchen2,@klightlm
@zechunliu,@balakrishnan_vr
@Fanyi_Xiao,@hyunwoojkim
@bilgeesra,@raghuraman,@vikasc
🚨VideoLLM from Meta!🚨
LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
📝Paper: huggingface.co/papers/2410.1…
🧑🏻💻Code: github.com/Vision-CAIR/LongV…
🚀Project (Demo): vision-cair.github.io/LongVU
We propose LongVU, a video LLM with a spatiotemporal adaptive compression mechanism designed for real-world hour-long video understanding. LongVU adaptively reduces the number of video tokens by leveraging (1) DINOv2 feature similarity across frames, (2) Cross-modal text-frame similarity, and (3) temporal frame similarity.
1. High quality on video-based QA: 67.6% on EgoSchema, 66.9% on MVBench, 65.4% on MLVU and 59.5% on VideoMME long
2. +5% accuracy boost on average across various video understanding benchmarks compared to LLaVA-OneVision and VideoChat2
3. Our edge model, LongVU-3B, also outperformed 4B counterparts such as VideoChat2(Phi-3) and Phi-3.5-vision-instruct by a large margin.
with: @xiaoqian_shen @liuzhuang1234 @Hu_Hsu @garvinchen2 @klightlm @zechunliu @balakrishnan_vr @Fanyi_Xiao @hyunwoojkim @bilgeesra @raghuraman @moElhoseiny @vikasc
KAUST Vision CAIR group retweeted
Join us tomorrow 8:30-1pm; panel will start 11:50am
Announcing our hybrid panel: Computer Vision in the Real World! 🌿
🗓️ 09/30 at 11:50 am CEST
📍 Attend in person at ECCV or virtually
Explore the challenges in real-world CV deployment for biodiversity and conservation. #AIforConservation #CV4E
KAUST Vision CAIR group retweeted
@hannah_kerner speaking now at the @CV4E_ECCV workshop on the future of geospatial foundation models for the earth 🌍🤩 – at Milan, Lombardy
