MultiFlyFour sensing modalities. One shared semantic view of the world.
A real-world aerial dataset connecting RGB, thermal, LiDAR, and radar through geometry-driven semantic label transfer.
Research Engineer · Fraunhofer IVI
Taehyoung Kim.
I’m a research engineer at Fraunhofer IVI working toward aerial autonomy. My work spans the path from data to deployment: collecting and curating aerial datasets, making multimodal annotation and training more efficient, and developing online perception and mapping alongside offline map reconstruction. I also work on getting these ideas running on real robots, bringing the models and software together into a working system.
The goal is to enable aerial robots to perceive and act, supporting path planning, navigation, and mission execution. I’m also exploring reinforcement learning (RL), vision-language navigation (VLN), and vision-language-action (VLA) models toward end-to-end autonomy, connecting observations and instructions to learned actions.
Here I share the research, experiments, and technical lessons that come out of that process.
Current focusMultimodal perception, mapping, and VLN/VLA for near-ground aerial autonomy.
01 / A little about me
From an idea
to a working system.
My research brings together video understanding, multimodal sensing, and semantic 3D mapping. I’m interested in connecting these ideas to perception systems that work beyond a benchmark.
This site is a place to share that process: the papers, the experiments, and the practical things I learn while building. It also leaves room for side projects and ideas outside my day-to-day research.
Read my bio02 / What I’ve been working on
Recent activity.
Recent papers and project updates. A small record of what I’m working on and sharing.
Publication
MultiFly preprint
A real-world aerial dataset connecting RGB, thermal, LiDAR, and radar through geometry-driven semantic label transfer. Kudos to Markus and Andreas.
Project update
ADAM project final presentation
Joined the ADAM project demo flights and presented our work on high-definition 3D semantic mapping. Kudos to Paul Spannaus and Markus Gross. Special thanks to the ADAM team.
Publication
VoxelFix preprint
Graph-based correction of semantic labels in completed voxel maps, keeping geometry and occupancy fixed. Kudos to Sunesh.
Publication
MissionBench preprint
Mission-level evaluation of multimodal language-model agents for aerial tasks, including qualitative failure analysis. Kudos to Suman and Jona.
03 / Research
Selected research.
VoxelFixBetter semantic maps. Same geometry.
Graph-based correction of semantic labels in completed voxel maps, keeping geometry and occupancy fixed.
SSPSeeing consistently, frame after frame.
Semantic similarity propagation improves temporal consistency in video semantic segmentation for autonomous flight.
MissionBenchFrom individual actions to complete missions.
Mission-level evaluation of multimodal language-model agents for aerial tasks, including qualitative failure analysis.
FlexPathLearning where a path could go.
Learned semantic path priors connect image understanding with image-based planning.
Applied research / Collaborative projects
From perception to flight.
VERUM · NiKoLaS · ADAM · ENGEL
Collaborative research, with full author credits on every paper.Full publication record
04 / The questions that connect it
Perceive. Represent.
Support decisions.
Three connected research directions. Distinct projects, shared questions about reliable robotic perception.
Real-time
perception
How can a robot understand what it sees, consistently over time and across sensing modalities?
Video perception / SSP3D scene
understanding
How can semantic geometry become a more reliable representation of the world?
Semantic mapping / VoxelFixAerial
autonomy
How can perception support planning, and how should we evaluate an agent’s complete mission?
Mission evaluation / MissionBench05 / Beyond the paper
Research meets
real systems.
My work sits at the intersection of computer vision, semantic mapping, and aerial robotics.
At Fraunhofer IVI, I work in a collaborative research environment where robotic perception connects with practical autonomous systems.
Explore collaborative projectsIdeas are better in conversation
Let’s connect.
See where it leads.
External PhD opportunities
I’m looking for an external PhD position in robotic perception and aerial autonomy. If your group has a potential fit, I’d love to talk.
Research & project collaboration
Have a research question or a project we could build together? I’m open to collaborations across perception, mapping, and autonomous systems.
Interesting opportunities
I’m always open to a conversation about roles and teams doing thoughtful, ambitious work. If you think we could build something exciting together, get in touch.
Beyond research
We don’t need a project to connect. I’m happy to meet new people, exchange ideas, or just have a good conversation. Say hello :)
Pick the channel that feels right for you.
I’d be happy to hear from you.