5 ms·

If there is something that isn't going to help the public perception of autonomous cars at all, it's releasing a compressed to shit capture of another video sho

If there is something that isn't going to help the public perception of autonomous cars at all, it's releasing a compressed to shit capture of another video showing a single camera angle from dozens. I would say it's a deliberate attempt to manipulate if I didn't also strongly believe ignorance on part of the police department has lead them to believe that autonomous cars could even exit a parking lot without data from many more than this one camera, not to mention the vastly more useful LIDAR on top. (That's before you consider the video angles shown here are just for dashcam purposes. The real cameras for the autonomous driving are in the sensor array on top of the roof)

jonathanyc 9y ago
Yeah. Does the Uber car really capture video at 480p and 15fps? Also only releasing the video conveniently ignores the fact that these cars have IR and LIDAR. The pedestrian is hard to see in this video essentially because it is dark and they are wearing dark clothing. Neither of these are at all obstacles to LIDAR and IR, and the video at least shows us that the road is clear of obstructions.
- jon_richards 9y ago
  >Does the Uber car really capture video at 480p and 15fps? Probably not, but it isn't like the inputs to the self-driving models really need to be better than that. Lower resolution helps your processing time a lot and there's little point in having an FPS higher than your processing time.
  - ClassyJacket 9y ago
    I'm sure at least the collision avoidance part of the system would need to poll at a much higher rate than 15fps. That's up to 67ms latency you're adding. With enough miles that delay could kill people.
    - jon_richards 9y ago
      Average human reaction speed is around 215ms. Not an apples-to-apples comparison because humans can react much faster to continuous situations (humans have a timing accuracy of around 9.5ms) while a machine-learning model is limited to only reacting once per frame, but still.
      - bonzini 9y ago
        Average human reaction speed when driving, from the time the dangerous situation happens to the time of first reaction, is usually considered to be in the 700-2000 ms range. The lower bound is with optimal lighting and after something has alerted the driver in advance (e.g. a police car dashing past you a moment before), and the first reaction is not "the brakes are slammed" but "the foot is lifted from the accelerator". Hitting the brakes can take another 100-200 ms.
      - Tobba_ 9y ago
        If you wan't to compare against human "sample rate", it'd be equivalent to at least ~200 FPS (in order to get the same accuracy with a camera). Sure, the signal takes a moment to plumb its way through, but that's irrelevant to spotting objects. If they're actually feeding data at 15 FPS into their ML model, then what the fuck were they expecting? Correlating movements at those framerates would be nigh-impossible. Relying on ML for this is already comically irresponsible, but that'd just be ridiculous.
        michaelcampbell 9y ago
        > If you wan't to compare against human "sample rate", it'd be equivalent to at least ~200 FPS Where does this figure come from?
        jon_richards 9y ago
        I think they're pulling it from noticeable monitor FPS rates. I'm not an expert on machine vision, so I couldn't tell you the FPS needed to correlate movements between frames.
frgtpsswrdlame 9y ago
Uber is more than welcome to release their video(s) themselves if this one represents an inaccurate view.
- stefan_ 9y ago
  Given there are no obstructions in the road, LIDAR (or even the decent cameras from the sensor array on top) would obviously show her on a clear collision path many seconds before impact. So I guess this is the ultimate litmus test if the culture at Uber has changed. They let this video stand and I don't personally think they should be allowed to drive another metre autonomously on public roads.
  - TillE 9y ago
    Any engineer on that project would have to be horrified that their system didn't prevent this accident. It failed, completely. I would be shocked if we see Uber resume testing any time soon, if at all.
    - brainwad 9y ago
      This is Uber we're talking about. They ran a red light while self-driving, shrugged it off and kept going. They had a team whose job was to avoid regulation by blacklisting the credit cards of police. I would be shocked if Uber does anything other than the bare minimum of apologising.
hcknwscommenter 9y ago
I think releasing much higher quality and higher dynamic range and multi-spectrum video is going to be a LOT more damming of driverless car tech. Just look at all the commentators here saying they would have not been able to stop in time based on this terrible camera footage (ignoring the vastly better capabilities of their own human eyes).
- stefan_ 9y ago
  I'm holding out hope that a good chunk of the public will understand this footage is not at all indicative of the vision capabilities of even the human eye, lest the actual sensor tech on that Uber car. Certainly a good number of people on Twitter do. If Waymo understood how to stick a knife in they would have replicated this scenario yesterday and released a video. Hell, you could take a Mercedes S class with night view assistant and it would spot this, no problem.
drcode 9y ago
Personally, I'm amazed that we get to see any video at all from this accident... Given that we live in an age where video is king, and given Uber's reputation for going to extreme lengths to get their way, I think we should be thankful to the Tempe Police for getting this important document to the public so quickly.
gkya 9y ago
> If there is something that isn't going to help the public perception of autonomous cars If there is something now that's going to help with the public perception of this cars that people claim to be autonomous, it's people like you just quitting trying to defend this and care about the deceased ATM. The software and hardware was sub-par, and the driver of a fucking TEST DRIVE was not looking at the road, let aside having hands ready on the steering wheel. The driver's feet were probably wandering around relaxing besides the pedals. TBH I don't give a flying fuck to the public image of autonomous (!) vehicles ATM.

jonathanyc 9y ago

Yeah. Does the Uber car really capture video at 480p and 15fps? Also only releasing the video conveniently ignores the fact that these cars have IR and LIDAR. The pedestrian is hard to see in this video essentially because it is dark and they are wearing dark clothing. Neither of these are at all obstacles to LIDAR and IR, and the video at least shows us that the road is clear of obstructions.

jon_richards 9y ago
>Does the Uber car really capture video at 480p and 15fps? Probably not, but it isn't like the inputs to the self-driving models really need to be better than that. Lower resolution helps your processing time a lot and there's little point in having an FPS higher than your processing time.
- ClassyJacket 9y ago
  I'm sure at least the collision avoidance part of the system would need to poll at a much higher rate than 15fps. That's up to 67ms latency you're adding. With enough miles that delay could kill people.
  - jon_richards 9y ago
    Average human reaction speed is around 215ms. Not an apples-to-apples comparison because humans can react much faster to continuous situations (humans have a timing accuracy of around 9.5ms) while a machine-learning model is limited to only reacting once per frame, but still.
    - bonzini 9y ago
      Average human reaction speed when driving, from the time the dangerous situation happens to the time of first reaction, is usually considered to be in the 700-2000 ms range. The lower bound is with optimal lighting and after something has alerted the driver in advance (e.g. a police car dashing past you a moment before), and the first reaction is not "the brakes are slammed" but "the foot is lifted from the accelerator". Hitting the brakes can take another 100-200 ms.
    - Tobba_ 9y ago
      If you wan't to compare against human "sample rate", it'd be equivalent to at least ~200 FPS (in order to get the same accuracy with a camera). Sure, the signal takes a moment to plumb its way through, but that's irrelevant to spotting objects. If they're actually feeding data at 15 FPS into their ML model, then what the fuck were they expecting? Correlating movements at those framerates would be nigh-impossible. Relying on ML for this is already comically irresponsible, but that'd just be ridiculous.
      - michaelcampbell 9y ago
        > If you wan't to compare against human "sample rate", it'd be equivalent to at least ~200 FPS Where does this figure come from?
        jon_richards 9y ago
        I think they're pulling it from noticeable monitor FPS rates. I'm not an expert on machine vision, so I couldn't tell you the FPS needed to correlate movements between frames.

frgtpsswrdlame 9y ago

Uber is more than welcome to release their video(s) themselves if this one represents an inaccurate view.

stefan_ 9y ago
Given there are no obstructions in the road, LIDAR (or even the decent cameras from the sensor array on top) would obviously show her on a clear collision path many seconds before impact. So I guess this is the ultimate litmus test if the culture at Uber has changed. They let this video stand and I don't personally think they should be allowed to drive another metre autonomously on public roads.
- TillE 9y ago
  Any engineer on that project would have to be horrified that their system didn't prevent this accident. It failed, completely. I would be shocked if we see Uber resume testing any time soon, if at all.
  - brainwad 9y ago
    This is Uber we're talking about. They ran a red light while self-driving, shrugged it off and kept going. They had a team whose job was to avoid regulation by blacklisting the credit cards of police. I would be shocked if Uber does anything other than the bare minimum of apologising.

hcknwscommenter 9y ago

I think releasing much higher quality and higher dynamic range and multi-spectrum video is going to be a LOT more damming of driverless car tech. Just look at all the commentators here saying they would have not been able to stop in time based on this terrible camera footage (ignoring the vastly better capabilities of their own human eyes).

stefan_ 9y ago
I'm holding out hope that a good chunk of the public will understand this footage is not at all indicative of the vision capabilities of even the human eye, lest the actual sensor tech on that Uber car. Certainly a good number of people on Twitter do. If Waymo understood how to stick a knife in they would have replicated this scenario yesterday and released a video. Hell, you could take a Mercedes S class with night view assistant and it would spot this, no problem.

drcode 9y ago

Personally, I'm amazed that we get to see any video at all from this accident... Given that we live in an age where video is king, and given Uber's reputation for going to extreme lengths to get their way, I think we should be thankful to the Tempe Police for getting this important document to the public so quickly.

gkya 9y ago

> If there is something that isn't going to help the public perception of autonomous cars If there is something now that's going to help with the public perception of this cars that people claim to be autonomous, it's people like you just quitting trying to defend this and care about the deceased ATM. The software and hardware was sub-par, and the driver of a fucking TEST DRIVE was not looking at the road, let aside having hands ready on the steering wheel. The driver's feet were probably wandering around relaxing besides the pedals. TBH I don't give a flying fuck to the public image of autonomous (!) vehicles ATM.