Improving YouTube video thumbnails with deep neural nets • Hacker News Board

Re: Improving YouTube video thumbnails with deep neural nets

#21

Post by baddox » Sat, Oct 10, 2015, 7:15 AM UTC

nacs wrote: post #7

Scaevolus wrote: post #2

I wonder if they included sexy images in their negative training sets -- many videos accrue millions of views (and ad dollars) by having a few frames of cleavage interspersed with other (often derivative) footage. It would be great if their algorithm picked a thumbnail that reflected the entire video, not just a few frames specifically chosen to game people's compulsive clicking.

Most of those are just manually selected thumbnails by the uploader. After uploading, YT gives you 3-5 thumbnails you can choose from. Also, partnered accounts are allowed to upload custom thumbnails (which can be any image, not necessarily even a screenshot from the video).

Not just partnered accounts. My YouTube account lets me upload custom thumbnails, and I'm certainly not followed. I have maybe a couple dozen videos with maybe a few hundred views among them all.

Re: Improving YouTube video thumbnails with deep neural nets

#22

Post by ZeroGravitas » Sat, Oct 10, 2015, 8:37 AM UTC

nacs wrote: post #7

Scaevolus wrote: post #2

I wonder if they included sexy images in their negative training sets -- many videos accrue millions of views (and ad dollars) by having a few frames of cleavage interspersed with other (often derivative) footage. It would be great if their algorithm picked a thumbnail that reflected the entire video, not just a few frames specifically chosen to game people's compulsive clicking.

Most of those are just manually selected thumbnails by the uploader. After uploading, YT gives you 3-5 thumbnails you can choose from. Also, partnered accounts are allowed to upload custom thumbnails (which can be any image, not necessarily even a screenshot from the video).

Is this definitely not algorithmic? I've been noticing for a while that videos might have an incidental flash of cleavage and then that is used as the thumbnail. I'd always wondered if this was arising "naturally" somehow (people pausing that scene perhaps?)

On the basis of the type if video I'd discounted manual intervention. Though if people can just upload any image I'm now surprised that they're not all like this.

Re: Improving YouTube video thumbnails with deep neural nets

#23

Post by lumpypua » Sat, Oct 10, 2015, 10:30 AM UTC

nacs wrote: post #14

JoshTriplett wrote: post #13

Earlier quoted context omitted.

> After uploading, YT gives you 3-5 thumbnails you can choose from. Can you pick an arbitrary video frame, or only one of the suggested thumbnails?

It automatically captures 3 different thumbnails (I guess using the algorithm in OP) and lets you select any 1 of those 3.

I presume they use the image selection as training data too—if not that seems like awfully low hanging data fruit.

Re: Improving YouTube video thumbnails with deep neural nets

#24

Post by mdpm » Sat, Oct 10, 2015, 11:33 AM UTC

Meanwhile, I still can't edit a playlist while playing it.

edit: constructively put - there's simpler stuff to fix UX and match user patterns still isn't there?

Re: Improving YouTube video thumbnails with deep neural nets

#25

Post by anjc » Sat, Oct 10, 2015, 1:43 PM UTC

Animats wrote: post #20

It looks like they prefer images with a few large faces near the center of the frame. That's probably the right answer for social media. (Plus a cat recognizer.) Used on news footage, you probably get the talking head rather than the news event.

We can't guess as to how the NN is preferring images, but it looks to me like it's preferring images with a high entropy in certain regions

Re: Improving YouTube video thumbnails with deep neural nets

#26

Post by GhotiFish » Sat, Oct 10, 2015, 6:58 PM UTC

mdpm wrote: post #24

Meanwhile, I still can't edit a playlist while playing it. edit: constructively put - there's simpler stuff to fix UX and match user patterns still isn't there?

When you have a big system, the most consistent argument against working on one thing is that you should be working on something else, this is true for everything in the system, because everyone has a different opinion on what that thing is.

For example: why should you spend time working on the playlist playback when youtube could instead spend time working on automatic categorization, content creators have to manually create playlists, even if they sequentially number their videos. Youtube shouldn't waste their time on playlist editing when it could be doing the right thing automatically.