Henghui Ding

Professor @FudanUni, woking on Computer Vision and Artificial Intelligence. Previously @ETH_en & @NTUsg

[Henghui Ding @HenghuiDing]

Introducing #WBench – the first comprehensive benchmark for interactive world models. It evaluates whether models can balance rendering, controllability, consistency, and physics across multi-turn interactions.

- Paper: https://t.co/O4hEHfKEBD
- Homepage: https://t.co/S2c58Y5frK

[Visual-Intelligence @VI_Journal_CSIG]

🎉🎉🎉Visual Intelligence is now in ESCI ! A huge thank you to our authors, reviewers, and readers for your support. We are excited to reach this major milestone!
CiteScoreTracker 2025 10.6
All papers since 2023 will be indexed!
https://t.co/AWU3d4mkaN

[Henghui Ding @HenghuiDing]

#CVPR2026 Introducing #EffectErase. It removes both the target object and its visual effects (e.g., shadows and reflections) from a video.

-Project: https://t.co/gek3N5pHYp
-Paper: https://t.co/zb5JhrhJaU
-Code: https://t.co/VIcZjcOV0h

[Henghui Ding @HenghuiDing]

#CVPR2026 Introducing #GlyphPrinter. It generates accurate glyphs even in the most challenging scenarios.

-Project: https://t.co/8UWehubYCC
-Paper: https://t.co/mpiSJ3vIGf
-Code: https://t.co/ROFWhBkKQw

[Tencent Hy @TencentHunyuan]

One static model does not fit all😭

We just dropped our latest work: Functional Neural Memory. Instead of static models, we generate custom "parameters" for every single input.

✅Prompt your model anytime
✅Instant personalization
✅Better instruction following
✅Flexible & dynamic memory (w/o memory bank✌️)

(🧵1/6)

[Visual-Intelligence @VI_Journal_CSIG]

🔥Check out a new method of image dehazing. The authors proposed a synergic expert modulation (SEM) mechanism to explicitly model context information.
🔗https://t.co/2GPc00zBVc
#dehaze
@SpringerEng

[Henghui Ding @HenghuiDing]

Excited to share that our paper “GREx: Generalized Referring Expression Segmentation, Comprehension, and Generation” is now published in IJCV! This work extends our CVPR 2023 Highlight paper GRES.

Project Page: https://t.co/O7AipBZ5RG

IJCV Version: https://t.co/OzVFUqW089

[Henghui Ding @HenghuiDing]

ICML Board Election 2025 is on! 🎉

I’m honored to be one of the candidates this year.
If you got the voting email, I’d love your support 🗳️🙏

Let’s keep building an inclusive and strong ML community!

#ICML2025 #BoardElection

[Visual-Intelligence @VI_Journal_CSIG]

📅Call for Papers: Special Issue on "Controllable Artificial Intelligence Visual Content Generation"📊💡
Submit your work on controllable AI-driven visual content generation for the SI!
Deadline: 31 August 2025
Deails: https://t.co/gEjyrJGMBm
#controllableAI #AIGC #CallForPapers

[Henghui Ding @HenghuiDing]

@deblinaforAI @eccvconf Waiting+1, a timeline would be helpful.

[Henghui Ding @HenghuiDing]

#CVPR2025 #PVUW The 4th Pixel-level Video Understanding in the Wild Challenge at CVPR 2025 @CVPR is open! Call for submission and challenge participation! Two challenge tracks: #MOSE track & #MeViS track. More details at https://t.co/aMcX3Lv2Hj

[Henghui Ding @HenghuiDing]

#ECCV2024 #LSVOS #MOSE #MeViS The 6th Large-scale Video Object Segmentation (LSVOS) Challenge at #ECCV2024 @eccvconf now! Thanks @nikhilaravi for sharing SAM2 on the workshop.

[Henghui Ding @HenghuiDing]

#ECCV2024 #LSVOS We are thrilled to introduce the 6th Large-scale Video Object Segmentation (LSVOS) Challenge at #ECCV2024 @eccvconf! 🚀 More challenging tracks and more topics!

All details have been released! Looking forward to participation!

Page: https://t.co/AI8BZDXC90

[Chen Change Loy @ccloy]

Excited to share that our survey paper "Transformer-Based Visual Segmentation: A Survey" is now accepted by TPAMI! 📝

🔍 Highlights:
1️⃣ Unlike previous surveys, we categorize transformer-based methods from a technical perspective.
2️⃣ We explore methods for mainstream tasks with DETR-like meta-architecture and related directions by tasks.
3️⃣ We re-benchmark representative works on image semantic & panoptic segmentation datasets.

Paper: https://t.co/GvpWhZJtnJ
GitHub repo: https://t.co/J8chltfCDg

Hard work from Xiangtai Li @xtl994 , Henghui Ding @HenghuiDing , Haobo Yuan @HarborYuan , Wenwei Zhang @wenweiz97 and other co-authors.

[Xiangtai Li @xtl994]

Finally, our transformer survey is accepted by T-PAMI-2024. Thanks for the co-authors' help. @HenghuiDing @liuziwei7 @ccloy The repo is at https://t.co/XDmGcqE4uC

Paper link: https://t.co/n1DR6NgYy4

[Henghui Ding @HenghuiDing]

#ECCV2024 #LSVOS We are thrilled to introduce the 6th Large-scale Video Object Segmentation (LSVOS) Challenge at #ECCV2024 @eccvconf! 🚀 More challenging tracks and more topics!

All details have been released! Looking forward to participation!

Page: https://t.co/AI8BZDXC90

[Henghui Ding @HenghuiDing]

Image Editing Survey!
ArXiv: https://t.co/qavWUrCgx1
Github: https://t.co/mr20nQ2aij

[Ziwei Liu @liuziwei7]

🤩Music to 3D Duet Dance Generation🤩

#ICLR2024 We propose 🕺Duolando💃, a GPT-based model that autoregressively predicts 3D motion for both leader and follower dancer @iclr_conf

- Project: https://t.co/IhFC609jBv
- Paper: https://t.co/OMQ0E9oBrc
- Code: https://t.co/IQuMR62KRI

[Henghui Ding @HenghuiDing]

@nikhilaravi @cgmsnoek

[Henghui Ding @HenghuiDing]

#CVPR2024 #PVUW Please consider submitting your work to our workshop on Pixel-level Video Understanding in the Wild (Track on Video Understanding) More details at https://t.co/QPThB6NElV

[#CVPR2026 @CVPR]

The list of #CVPR2024 workshops is now available!

https://t.co/vj7uSeBYqT

[Henghui Ding @HenghuiDing]

@xtl994 @HarborYuan The paper is Open Access here: https://t.co/Wnz69sZaK2