I've suddenly realized that computer vision of real images is very much solvable and that it is now just a matter of engineering. I was so stuck before because you can't make the simple assumptions in screenshot computer vision that you can in real computer vision. This makes experience probably necessary to effectively learn from screenshots. Objects in real images to not change drastically in appearance, position or other dimensions in unpredictable ways.
The reason I came to the conclusion that it's a lot easier than I thought is that I found a way to describe why existing solutions work, how they work and how to come up with even better solutions. I've also realized that I don't actually have to implement it, which is what is most difficult because even if you know a solution to part of the problem has certain properties and issues, implementing it takes a lot of time. Whereas I can just assume I have a less than perfect solution with the properties I predict from other experiments. Then I can solve the problem without actually implementing every last detail. *First*, existing methods find observations that are likely true by themselves. They find data patterns that are very unlikely to occur by coincidence, such as many features moving together over several frames of a video and over a statistically significant distance. They use thresholds to ensure that the observed changes are likely transformations of the original property observed or to ensure the statistical significance of an observation. These are highly likely true observations and not coincidences or noise. *Second*, they make sure that the other possible explanations of the observations are very unlikely. This is usually done using a threshold, and a second difference threshold from the first match to the second match. This makes sure that second best matches are much farther away than the best match. This is important because it's not enough to find a very likely match if there are 1000 very likely matches. You have to be able to show that the other matches are very unlikely, otherwise the specific match you pick may be just a tiny bit better than the others, and the confidence of that match would be very low. So, my initial design plans are as follows. Note: I will probably not actually implement the system because the engineering part dominates the time. I'd rather convert real videos to pseudo test cases or simulation test cases and then write a psuedo design and algorithm that can solve it. This would show that it works without actually spending the time needed to implement it. It's more important for me to prove it works and show what it can do than to actually do it. If I can prove it, there will be sufficient motivation for others to do it with more resources and man power than I have at my disposal. *My Design* *First, we use high speed cameras and lidar systems to gather sufficient data with very low uncertainty because the changes possible between data points can be assumed to be very low, allowing our thresholds to be much smaller, which eliminates many possible errors and ambiguities. *Second*, *we have to gain experience from high confidence observations. These are gathered as follows: 1) Describe allowable transformations(thresholds) and what they mean. Such as the change in size and position of an object based on the frame rate of a camera. Another might be allowable change in hue and contrast because of lighting changes. With a high frame rate camera, if you can find a match that is within these high confidence thresholds in multiple dimensions(size, position, color, etc), then you have a high confidence match. 2) Find data patterns that are very unlikely to occur by coincidence, such as many features moving together over several frames of a video and over a statistically significant distance. These are highly likely true observations and not coincidences or noise. 3) Most importantly, make sure the matches we find are highly likely on their own and unlikely to be coincidental. 4) Second most importantly, make sure that any other possible matches or alternative explanations are very unlikely in terms of distance( measured in multiple dimensions and weighted by the certainty of those observations). These should also be in terms of the thresholds we used previously because those define acceptable changes in a normalized way. *That is a rough description of the idea. Basically highly likely matches and very unlikely for the matches to be incorrect, coincidental or mistmatched. * Third, We use experience, when we have it, in combination with the algorithm I just described. If we can find unlikely coincidences between our experience and our raw sensory observations, we can use this to look specifically for those important observations the experience predicts and verify them, which will in turn give us higher confidence of inferences. Once we have solved the correspondence problem like this, we can perform higher reasoning and learning. Dave ------------------------------------------- agi Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/ Modify Your Subscription: https://www.listbox.com/member/?member_id=8660244&id_secret=8660244-6e7fb59c Powered by Listbox: http://www.listbox.com
