@lix709

Assistant Professor at University of Reading

 Reading, England
xiangli.ac.cn
Joined December 2022
  • Tweets10
  • Following47
  • Followers45
RT @Tianjinhuang1: 🚀 Excited to launch our REOBench — the first comprehensive benchmark for evaluating the robustness of Earth Observation…

REOBench: Benchmarking Robustness of Earth Observation Foundation Models

Earth observation foundation models have shown strong generalization across multiple Earth observation tasks, but their robustness under real-world perturbations remains underexplored. To bridge...

arxiv.org
1
25
📢 We have extended our archival papers submission deadline to March 24th! If you have any 3D project fresh-off-the-hook, consider submitting to our workshop 🧐 bit.ly/4bYALHE @CVPR #CVPR2024
Habib Slim@habib__slim4 Mar 2024
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2024 @CVPR 🏆 Challenges: bit.ly/4a2b7A5 📢 Call for Papers: bit.ly/4bYALHE C3DV will host two challenges this year, and an amazing set of speakers! 🧵👇
5
1
14
4,058
Join us for the Workshop on Compositional 3D Vision (C3DV) #CVPR2024 @CVPR 🏆 Challenges: bit.ly/4a2b7A5 📢 Call for Papers: bit.ly/4bYALHE C3DV will host two challenges this year, and an amazing set of speakers! 🧵👇
1
8
2
59
12,643
🚀Excited to share our MiniGPT-v2. It can perform many complex vision-language and visual grounding  tasks with simple interaction 🔗Demo & Project: minigpt-v2.github.io/ @tikgiau @xiaoqian_shen @lix709 @zechunliu @PengchuanZ  Raghuraman @vikasc @YoungXiong1 @moElhoseiny
6
57
9
208
195,429
Replying to @elluba
@elluba kicks off the second day of the SAAI AI Art History symposium. Livestream: youtube.com/watch?v=gyKEAMeO… Program: cemse.kaust.edu.sa/events/ev…
1
5
4
1,131
Thrilled to announce the release of 3DCoMPaT++: a richly annotated, multimodal 2D/3D dataset of more than 10 million stylized 3D shapes!🎉 🏆Challenge @CVPR 2023: bit.ly/43T1IIP 📚Doc: bit.ly/3N0OHGV 💻Code: bit.ly/3ou4aoK 🧵👇
2
48
4
199
37,371
Xiang Li retweeted
Deyao Zhu
@tikgiau17 Apr 2023
🚀 Excited to share MiniGPT-4, an open-sourced model performing complex vision-language tasks like GPT-4! ✍️ Write poems for photos 🔍 Pinpoint problems & offers solutions 🌐 Draft ➡️ Website and more! 🔗 @Gradio Demo & Project: minigpt-4.github.io/
39
236
45
960
280,627
We are excited to share our new work, Video ChatCaptioner, which can generate the enriched video spatiotemporal description through the conversation between ChatGPT and BLIP-2. Preprint: arxiv.org/abs/2304.04227 Code: github.com/Vision-CAIR/ChatC… @tikgiau @KAUSTVisionCAIR
3
40
4
170
29,763
Xiang Li retweeted
AK
@_akhaliq3 Feb 2023
SceneDreamer: Unbounded 3D Scene Generation from 2D Image Collections abs: arxiv.org/abs/2302.01330 project page: scene-dreamer.github.io/
10
168
30
699
317,658
Replying to @CVPR
@CVPR accepted the paper (code released). MoStGAN-V: Video Generation with Temporal Motion Styles @xiaoqian_shen, Xiang Li, Mohamed Elhoseiny paper: arxiv.org/abs/2304.02777 Website,Code,Videos: xiaoqian-shen.github.io/MoSt… Congrats, @xiaoqian_shen on her first CVPR paper!
3
3
1
14
1,999
Load more