chenguoz / pointscnet

Code Release of PointSCNet: Point Cloud Structure and Correlation Learning based on Space Filling Curve guided Sampling

👋 Hi there!

I’m Xingye Chen, an enthusiastic Algorithm Engineer on a mission to revolutionize the realms of AI, computer vision, and machine learning. My journey in technology is fueled by an insatiable curiosity and a relentless drive to explore the uncharted territories of innovation.

🌟 Current Endeavors

🚀 Pushing Boundaries: Crafting cutting-edge algorithms for Few-shot Learning, 3D Point Cloud Learning, and Visual Language Models.
🔬 Researching at Huazhong University of Science and Technology: Delving deep into the frontiers of AI technologies to uncover new possibilities.

🤝 Seeking Collaboration

🌐 Join Forces: Let's team up on groundbreaking research projects in computer vision, deep learning, and AI-driven solutions.
🌍 Real-World Impact: I’m keen on innovative applications that harness AI for transformative real-world impact, especially in unsupervised learning and 3D shape analysis.

🏅 Honors & Accolades

• 🥇 Huawei ICT Competition Global Finals: Global First Prize, 2022
• 🥇 Huawei China University ICT Competition National Finals: National First Prize, 2021
• 🏅 CCF Big Data and Computational Intelligence Competition: National 15th Place, 2021
• 💡 Huawei Research Innovation Scholarship, 2022
• 🎓 President’s Scholarship, China University of Geosciences, 2020

🌐 Let's Connect

Feel free to dive into my projects and reach out if you’re interested in collaboration or just want to chat about the latest advancements in AI and machine learning. Together, we can push the boundaries of what’s possible!

pointscnet's Issues

Train on Custom Dataset

Hello, awesome paper! Will this code be modified so we can train it on a custom dataset? Or will it be available as a network in another repository.

Is any bug in Spatial Attention?

文中提到的Spatial Attention，如下图所示，池化貌似在点云数量维度进行的，但是代码是在通道进行的，只是将输出维度设置为点云的数量，请问是bug吗？

主页上那个效果图是怎么画的呀