Design deep learning models for visual navigation,VLN,and outdoor visual place recognition
Develop multimodal perception pipelines combining RGB,depth,segmentation,and language navigation
Research scene understanding for pedestrian-dense streets,repetitive facades,and GNSS-degraded corridors
What they want
Hold a PhD in Computer Science,Robotics,AI,Computer Vision,or a related field
Show strong publications or research in visual navigation,VLN,embodied AI,or robot perception
Demonstrate expertise in at least two listed areas such as VLN,VPR,perception,or RL navigation
Advertisement
Nice to have
Deploy learned navigation models on physical robot platforms,not simulation-only experience
Apply LiDAR-visual fusion or hybrid metric-topological navigation approaches in practice
Use large-scale vision-language models for robotic instruction following in navigation tasks
Seoulstart's read
Work authorization
Not stated, verify with recruiter
Language beyond English
No language beyond English stated
Context
Onsite One-North role; the posting does not state work-pass sponsorship eligibility.
This is Seoulstart's analysis of the public job description, not the employer's stated policy. Verify work-pass eligibility, language requirements, and remote allowances directly with the recruiter.