Show HN: API for detecting people, cars, and everyday objects in images • Hacker News Board

Re: Show HN: API for detecting people, cars, and everyday objects in images

#71

Post by apu » Sun, Jan 13, 2013, 3:57 PM UTC

EwanG wrote: post #69

apu wrote: post #48

Looks like this is using training data from the PASCAL VOC object detection challenge [1], which is the standard benchmark for evaluating object detection performance in computer vision. Object detection is an extremely tough problem (some would say it is the computer vision problem ;-)), and while we've made a lot of progress in the past decade, the best methods are still terrible [2] -- average detection precision…

One of my main hobbies is photography. I do mainly outdoor shots, and really enjoy macros of flowers. The problem being that "oh last weekend I took an amazing shot of a purple flower" isn't all that helpful for someone who is trying to find a picture of an iris. When someone comes up with an algorithm that can take my shot, compare it to a library, and tell me what wildflower it is, I will be a happy camper. I suspe…

It's not quite what you want, but I worked on Leafsnap [1], which automatically identifies trees by their leaves, using computer vision techniques. We focused on leaves since they are present throughout much more of the year than flowers. Our free apps also include high-resolution, high-quality photos of all aspects of the species we cover -- leaves, flowers, fruits, bark, etc. So you can at least browse through and compare the flowers you're looking at with those in the app.

Our current coverage is of the trees of the northeast US (about 200 species), but we are working on expanding that.

[1] http://leafsnap.com

Re: Show HN: API for detecting people, cars, and everyday objects in images

#72

Post by apu » Sun, Jan 13, 2013, 4:00 PM UTC

eloisius wrote: post #57

apu wrote: post #48

Looks like this is using training data from the PASCAL VOC object detection challenge [1], which is the standard benchmark for evaluating object detection performance in computer vision. Object detection is an extremely tough problem (some would say it is the computer vision problem ;-)), and while we've made a lot of progress in the past decade, the best methods are still terrible [2] -- average detection precision…

Isn't the 30-50% only applicable to doing object recognition? I.e. multi-classification. In this case, you have to tell it which object you're looking for.

The relevant table on the results page is Table 3, which is detection performance. Classification is actually an easier problem (see Table 1), in part because the types of scenes in which different classes appear are often quite different, making it easy to avoid some "easy" mistakes.

Re: Show HN: API for detecting people, cars, and everyday objects in images

#73

Post by tripzilch » Mon, Jan 14, 2013, 4:19 PM UTC

martininmelb wrote: post #58

cynwoody wrote: post #54

Earlier quoted context omitted.

It got zero of six airplanes for the link below, even though the images are not overlapping and are against a blue sky background: http://www.keithcarter.com/wp-content/uploads/2009/10/blue-a...

But it did find one potted plant for that image. I could not see it (bottom left hand corner).

"Not again."

Re: Show HN: API for detecting people, cars, and everyday objects in images

#74

Post by tripzilch » Mon, Jan 14, 2013, 4:24 PM UTC

jluan wrote: post #24

Hey everybody, OP here. Thanks for the great feedback! We're really happy that so many people have checked this out. One thing that I want to mention: our service was built favoring Precision over Recall; we reasoned that we'd rather have a low number of false positives and make sure that when we do report a detection, that it actually is one. Thus, our service may occasionally miss instances. I'm going to implement…

This is a Dutch street, therefore it has many bikes in it: http://i.imgur.com/qQwAS.jpg .

Your application detects none of them... Is it because my ancient phone camera's pics are too grainy? Or do the bikes need to be en profile to be detected properly? Or maybe it's trained to detect bikes with people on them, instead of bikes parked in the street?