DyRAD optimizes a set of zero-extent point reflectors against recorded radar measurements.
Radar rendering that captures Doppler in dynamic scenes by separating sensor blur from geometry
Keidar et al. introduce DyRAD, which models driving scenes as static and tracked point reflectors and renders full range-azimuth-Doppler tensors through a fixed analytic point-spread function.
Big Tech
Merav Keidar · Tomer Borreda · Rajalakshmi Nandakumar · Or Litany
Technion · Cornell Tech · NVIDIA
Research Digest··3 min read
Keidar et al.
Why this paper
From NVIDIA and 2 others
In one line
Dynamic driving radar scenes can be reconstructed and re-rendered as range-azimuth-Doppler tensors from novel viewpoints using tracked point reflectors and a fixed analytic point-spread function.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors (3 noted)
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§