NVIDIA / NVIDIA/ncore

Per sensor timing offsets

Open
#136 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
197
Forks
35
Avg merge
15h 8m
Merged PRs (30d)
18

Description

Problem Statement

Thank you very much for this library!

I’d like to know if you have added to plan to add the ability to add per sensor timing offsets (shifting essentially all capture timestamps of a given sensor at once).
We work with data collected on different devices and sometimes there is a timing offsets between sensors, which needs to be corrected offline sometimes.
It would be really useful if the NCORE file can be written with original timestamps (convenience), while offering the ability to correct/adjust the relative time offsets between sensors when reading data from NCORE (or via a separate calibration record).
Practical usecase could be projecting lidar onto cameras and adding a time offset to the lidar of +/- N ms and visually or algorithmically inspect the alignment.

Proposed Solution

Either add the ability in the runtime evaluation APIs to add a time offset when evaluating sensor poses or add a dedicated record (similar to intrinsics) with timing offsets, which will be applied when evaluating frame times.

Alternatives Considered

No response

Additional Context

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the runtime evaluation APIs and the NCORE reading path mentioned in the issue; no files or tests are named. Compare the proposed runtime offset and dedicated calibration-record approaches, and define completion as applying a per-sensor timing adjustment while preserving original timestamps in written NCORE files.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
audio-video-rtc, backend-api-design, robotics
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.