PhD student @cmuhcii | Prev @berkeley_ai, @Apple Robotics, @Meta @RealityLabs, @hcikaist. 🦾 Making sense of sensing, for people!
Joined April 2018
- Tweets106
- Following912
- Followers1.3K
- Likes2.6K
Pinned Tweet
What if the XR glasses could place a microphone on your fingertip without actually attaching anything to your hand?
In our recent #CHI2026 work, SoundBubble captures subtle interaction sounds inside a small acoustic “bubble” around the finger, enabling a range of new interaction techniques.
Digital twins for robots and dynamics! RoboSimStudio is a browser-based toolkit that makes simulation more interactive and accessible for both users and agents.
We have the registration live! Hope to see you in Detroit 🔥
đź“ŁRegistration for #UIST2026 is now open! Join us in Detroit for groundbreaking research, mind-blowing demos, and incredible networking.
đź’ˇEarly-bird rates are available by registering before September 4th!
Register now: cvent.me/zd9527
More info: uist.acm.org/2026/attending/…
Graph-as-Policy (GaP) creates computation graphs (like ROS) for more modular and interpretable robot execution, yet generalizable to various tasks.
This is my first collaboration at UC Berkeley AUTOLAB as a visiting researcher. Shout out to lead authors Eric and Shuangyu, and I am glad I could contribute to this!
Graph-as-Policy (GaP) is a new variant of Agentic Robotics from @NVIDIA and @UCBerkeley that builds computation graphs (like ROS) to ensure modularity, manage complexity, and facilitate interpretability.
🧵Open code and paper: graph-robots.github.io/gap/
#Robotics #Automation #AgenticAI #EmbodiedAI
Check out @SCSatCMU’s wonderful feature on our recent project, SoundBubble!
SoundBubble helps isolate and amplify the sounds of fingers as they tap, trace or rap various surfaces, or use different tools. That information could give computers important hints about what users are doing and how the computer might help them.
cs.cmu.edu/news/2026/listeni…
Last but not least, I’ll be chairing the “Inferring Human State” session at #CHI2026. There are many exciting projects from amazing authors on sensing-based human pose estimation and psychological state inference. Join us in Room 127 at 9:00 AM.
For more details, check out the paper and source code:
* paper: daehwakim.com/soundbubble/
* code: github.com/FIGLAB/SoundBubbl…
And you can watch the full video here! youtu.be/UDyzusbP8C8
Daehwa Kim retweeted
🔋 What if wearables never needed batteries?
Power-over-Skin research project by @oldestasian, @talking_kim & @hciprof from @CarnegieMellon uses intra-body RF energy to power battery-free wearables—even head-to-toe and through clothing.
In the demo, a controller ring built with our XIAO nRF52840 runs sensing & BLE on just 61–1500 µW, showing what ultra-low-power MCUs can unlock for always-on wearables.
Watch the video by the Future Interfaces Group 👇
tinyurl.com/2y3fc96b
Daehwa Kim retweeted
Good Old-Fashioned Engineering Can Close the 100,000-Year “Data Gap” in Robotics. Science Robotics Editorial:
science.org/doi/10.1126/scir…
Daehwa Kim retweeted
đź“Ł The Student Volunteer lottery for #UIST2025 is now open! Be part of the team that brings UIST to life and make valuable connections with fellow students and leading researchers.
🔗 Application details: uist.acm.org/2025/cfp/#stude…
🗓️ Deadline: August 24, 2025 (AoE)
Yes, your iPhone already has a projector—using invisible light! 📱✨ In my latest research at @acm_chi, we repurpose its LiDAR dot pattern as a visual marker, enabling paired AR experiences without any online communication or pre-registration.
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
Research is done in amazing collaboration with @nneonneo @hciprof ✨ You can find more details from our recently published paper at #CHI2025. Here are the preprint, full video, and code: figlab.com/research/2025/pat…
