Lightnews — Scholar-powered news

Zhaochong An

@zhaochongan.bsky.social

150 followers 170 following 10 posts

PhD student at University of Copenhagen🇩🇰 Member of @belongielab.org, ELLIS @ellis.eu, and Pioneer Centre for AI🤖 Computer Vision | Multimodality MSc CS at ETH Zurich 🔗: zhaochongan.github.io/

Posts Media Videos Starter Packs

Pinned

Zhaochong An @zhaochongan.bsky.social · Apr 23

I will present our #ICLR2025 Spotlight paper MM-FSS this week in Singapore!

Curious how MULTIMODALITY can enhance FEW-SHOT 3D SEGMENTATION WITHOUT any additional cost? Come chat with us at the poster session — always happy to connect!🤝

🗓️ Fri 25 Apr, 3 - 5:30 pm
📍 Hall 3 + Hall 2B #504

More follow

Zhaochong An @zhaochongan.bsky.social · Apr 23

📜 Paper: Multimodality Helps Few-shot 3D Point Cloud Semantic Segmentation (arxiv.org/pdf/2410.22489)
🔗 Code: github.com/ZhaochongAn/...

#ICLR2025 #Multimodality #3DSegmentation @belongielab.org @ellis.eu @ethzurich.bsky.social @ox.ac.uk

Zhaochong An @zhaochongan.bsky.social · Apr 23

I will present our #ICLR2025 Spotlight paper MM-FSS this week in Singapore!

Curious how MULTIMODALITY can enhance FEW-SHOT 3D SEGMENTATION WITHOUT any additional cost? Come chat with us at the poster session — always happy to connect!🤝

🗓️ Fri 25 Apr, 3 - 5:30 pm
📍 Hall 3 + Hall 2B #504

More follow

Zhaochong An @zhaochongan.bsky.social · Feb 11

7/ For a deep dive into our model design, modality analysis, and experiments, check out our paper and code here.

📝: arxiv.org/pdf/2410.22489
💻: github.com/ZhaochongAn/Multimodality-3D-Few-Shot

Zhaochong An @zhaochongan.bsky.social · Feb 11

6/ The result? New SOTA few-shot performance that opens up exciting possibilities for multimodal adaptations in robotics, personalization, virtual reality, and more! ☀️

Zhaochong An @zhaochongan.bsky.social · Feb 11

5/ By fusing these modalities, MM-FSS generalizes to novel classes more effectively—even when the 3D-only connection between support and query is weak. 🚀

Zhaochong An @zhaochongan.bsky.social · Feb 11

4/ On the model side, our multimodal fusion designs harness cross-modal complementary knowledge to boost novel class learning, and test-time cross-modality calibration mitigates training bias.

Zhaochong An @zhaochongan.bsky.social · Feb 11

3/ That’s where MM-FSS comes in! We introduce two commonly overlooked modalities:
💠 2D images (leveraged implicitly during pretraining)
💠 Text (using class names)

—all at no extra cost beyond the 3D-only setup. ✨

Zhaochong An @zhaochongan.bsky.social · Feb 11

2/ Previous methods rely solely on a single modality—few-shot 3D support samples—to learn novel class knowledge.

However, when support and query objects look very different, performance can suffer, limiting effective few-shot adaptation. 🙁

Zhaochong An @zhaochongan.bsky.social · Feb 11

1/ Our previous work, COSeg, showed that explicitly modeling support–query relationships via correlation optimization can achieve SOTA 3D few-shot segmentation.

With MM-FSS, we take it even further!

Ref: COSeg arxiv.org/pdf/2410.22489

Zhaochong An @zhaochongan.bsky.social · Feb 11

Thrilled to announce "Multimodality Helps Few-shot 3D Point Cloud Semantic Segmentation" is accepted as a Spotlight (5%) at #ICLR2025!

Our model MM-FSS leverages 3D, 2D, & text modalities for robust few-shot 3D segmentation—all without extra labeling cost. 🤩

arxiv.org/pdf/2410.22489

More details👇

Reposted by Zhaochong An

Belongie Lab @belongielab.org · Nov 17

Logging on! 🧑‍💻🦋 We're the Belongie Lab led by @sergebelongie.bsky.social. We study Computer Vision and Machine Learning, located at the University of Copenhagen and Pioneer Centre for AI. Follow along to hear about our research past and present! www.belongielab.org

Belongie Lab - Home

Belongie Lab -- Home.

www.belongielab.org