Camera based distance to detected object including cars and pedestrians

Please provide the following info:
Hardware Platform: [Example: DRIVE AGX Xavier™ Developer Kit]
Software Version: [Example: DRIVE Software 10]
Host Machine Version: [native Ubuntu 18.04]
SDK Manager Version: []

Website mentions the following

The NVIDIA DRIVE™ Perception pipeline for obstacle perception consists of interacting algorithmic modules built around NVIDIA DRIVE™ Networks DNNs, along with DNN post-processing. Capabilities include:

*** Obstacle detection and classification, including cars and pedestrians, as well as distance to object detection (based on DriveNet DNN)**

  • Drivable free-space detection (based on OpenRoadNet DNN)
  • Camera image clarity detection and classification (based on ClearSightNet DNN)
  • Semantic motion segmentation (SMS) for detection of both static and dynamic objects

Also, there is a blog -DRIVE Labs: Detecting the Distance | NVIDIA Blog
which mentions about “distance-to-object detection using data from a single camera”

Can you please let me know the sample code where this is implemented?

Drivenet has object dwObjectCamera which has below attribute. However, value is always 0.

// Kinematics
/// Radial distance (in meter) from host car to object
float32_t distance_m;

Thanks & Regards,

Hi @hitesh.pandya1,

The distance information can be enabled via hasObjectDepth but there are some issues.
You need to wait for the next DRIVE Software release. The detail is in below topic.

OK, Thanks for the reply. Please let us know the release schedule.

For DRIVE Software 11 schedule, please contact with your nvidia representive.
Sorry for any inconvenience!

Hello @VickNV,

Are these issues solved? If not… Do you have a date for the DRIVE Software release?


1 Like

Hi @daniel.yustegalvez ,

Please contact with your nvidia rep for the schedule question. Sorry for any inconvenience.