4 ms·

So this has got to be prohibitively expensive for home gamers if the request for information form's minimum order size listed is <100. Is there anything compar

So this has got to be prohibitively expensive for home gamers if the request for information form's minimum order size listed is <100. Is there anything comparable for hobby project lidar scanning? Not necessarily wide angle, since you can always use a turntable, but capable of generating moderately dense point clouds.

verdverm 6y ago
The latest Azure Kinect has a megapixel depth sensor, costd $400 iirc Video is 4 MP lots of nice open source to go with it these days
shusson 6y ago
> minimum order size listed is <100 not to mention they don't even specify how you can power or connect to the sensor.
- azernik 6y ago
  Usually that's in the datasheet they send during initial negotiations.
  - malcolmwhite 6y ago
    In my (coworker's) experience with this sensor, they don't send a datasheet or give any specs on power, but they will replace your demo unit when you ruin your first one by trying to power it with PoE.
    - azernik 6y ago
      headdesk Fucking hardware vendors
jonas21 6y ago
If you have one of the new iPhones with lidar, there are a few apps (like Scaniverse [1]) that can generate dense point clouds and textured meshes. [1] https://scaniverse.com https://scaniverse.com
- dogma1138 6y ago
  I personally can’t wait till iPhone sensors come on a package that can be easily connected to a Pi or a Jetson these will leak into the generic supply chain in 1-2 refreshes.
  - samatman 6y ago
    Hope so but this depends on how patent encumbered they are, right? Like maybe you can just buy the part off Aliexpress but if no one is willing to put it onto a breakout board, let alone into a product, that's less broadly useful.
    - dogma1138 6y ago
      Chinese companies don’t care much for patents, you already have things like the HQ Pi camera with a decent Sony sensor simply because those are now more readily available. I don’t think the module itself is protected by patents that would prevent breakout boards especially from Chinese companies, you can buy replacement parts that Apple claims are counterfeit but in reality they are identical to the originals because they come from the same supplier just not from the same supply chain. The only question is how easy would it be to connect those sensors and make use of them if a lot of the logic that is required to drive them is buried in the Ax SoC from Apple it would be difficult if they use a rather standard protocol then it would rather simple.
      - mike_d 6y ago
        The Chinese government cares deeply about patents for companies that drive significant GDP.
        dogma1138 6y ago
        Which is why you can buy so many apple parts on the grey market in China already? Either Apple doesn’t care, the Chinese government doesn’t care, or they cannot stop the sellers and likely a combination of all of the above.
dljsjr 6y ago
Typically the LIDARs that are on autonomous vehicles are pretty low "resolution" compared to smaller LIDAR packages like a Hokuyo or depth sensors like a Kinect. It wouldn't be good for fine detail scanning.
- echelon 6y ago
  Azure Kinect isn't great at resolution, either. Its depth sensor is an embarrassingly small VGA resolution, which sucks for realtime photogrammetry. Point cloud scenes look like sludge. Are there any "real time" depth sensors with high spatial and sensor resolution? I've been looking at FGPA-based solutions like Nerian Scarlett, and this seems to be closer to what I want [1], but still not quite good enough. [1] https://nerian.com/products/scarlet-3d-depth-camera/ https://nerian.com/products/scarlet-3d-depth-camera/
  - dljsjr 6y ago
    The reason I put resolution in quotes is that I wasn't referring to the image/buffer resolution but rather the ability to resolve fine details. For example, Hokuyo LIDARs can resolve details w/ ~millimeter resolution at close range (less than 10m). Another way to think about it is if you want to map your point cloud as a set of voxels in an octree, the smallest voxel is about 1m x 1m x 1m. Conversely, large LIDARs for vehicles like Ibeo and Velodyne and such typically map things with a resolution in the range of centimeters or larger. When you're looking for large things like other vehicles, people, obstacles, etc. then you can make those tradeoffs in the sensor design and fall back on things like vision for small feature detection in a more cost effective way.

verdverm 6y ago

The latest Azure Kinect has a megapixel depth sensor, costd $400 iirc Video is 4 MP lots of nice open source to go with it these days

shusson 6y ago

> minimum order size listed is <100 not to mention they don't even specify how you can power or connect to the sensor.

azernik 6y ago
Usually that's in the datasheet they send during initial negotiations.
- malcolmwhite 6y ago
  In my (coworker's) experience with this sensor, they don't send a datasheet or give any specs on power, but they will replace your demo unit when you ruin your first one by trying to power it with PoE.
  - azernik 6y ago
    headdesk Fucking hardware vendors

jonas21 6y ago

If you have one of the new iPhones with lidar, there are a few apps (like Scaniverse [1]) that can generate dense point clouds and textured meshes. [1] https://scaniverse.com https://scaniverse.com

dogma1138 6y ago
I personally can’t wait till iPhone sensors come on a package that can be easily connected to a Pi or a Jetson these will leak into the generic supply chain in 1-2 refreshes.
- samatman 6y ago
  Hope so but this depends on how patent encumbered they are, right? Like maybe you can just buy the part off Aliexpress but if no one is willing to put it onto a breakout board, let alone into a product, that's less broadly useful.
  - dogma1138 6y ago
    Chinese companies don’t care much for patents, you already have things like the HQ Pi camera with a decent Sony sensor simply because those are now more readily available. I don’t think the module itself is protected by patents that would prevent breakout boards especially from Chinese companies, you can buy replacement parts that Apple claims are counterfeit but in reality they are identical to the originals because they come from the same supplier just not from the same supply chain. The only question is how easy would it be to connect those sensors and make use of them if a lot of the logic that is required to drive them is buried in the Ax SoC from Apple it would be difficult if they use a rather standard protocol then it would rather simple.
    - mike_d 6y ago
      The Chinese government cares deeply about patents for companies that drive significant GDP.
      - dogma1138 6y ago
        Which is why you can buy so many apple parts on the grey market in China already? Either Apple doesn’t care, the Chinese government doesn’t care, or they cannot stop the sellers and likely a combination of all of the above.

dljsjr 6y ago

Typically the LIDARs that are on autonomous vehicles are pretty low "resolution" compared to smaller LIDAR packages like a Hokuyo or depth sensors like a Kinect. It wouldn't be good for fine detail scanning.

echelon 6y ago
Azure Kinect isn't great at resolution, either. Its depth sensor is an embarrassingly small VGA resolution, which sucks for realtime photogrammetry. Point cloud scenes look like sludge. Are there any "real time" depth sensors with high spatial and sensor resolution? I've been looking at FGPA-based solutions like Nerian Scarlett, and this seems to be closer to what I want [1], but still not quite good enough. [1] https://nerian.com/products/scarlet-3d-depth-camera/ https://nerian.com/products/scarlet-3d-depth-camera/
- dljsjr 6y ago
  The reason I put resolution in quotes is that I wasn't referring to the image/buffer resolution but rather the ability to resolve fine details. For example, Hokuyo LIDARs can resolve details w/ ~millimeter resolution at close range (less than 10m). Another way to think about it is if you want to map your point cloud as a set of voxels in an octree, the smallest voxel is about 1m x 1m x 1m. Conversely, large LIDARs for vehicles like Ibeo and Velodyne and such typically map things with a resolution in the range of centimeters or larger. When you're looking for large things like other vehicles, people, obstacles, etc. then you can make those tradeoffs in the sensor design and fall back on things like vision for small feature detection in a more cost effective way.