4 ms·

This is a demonstration of the abilities of OpenAI's CLIP neural network. The tool will download a YouTube video, extract frames at regular intervals and precom

by vladoh 6y ago

This is a demonstration of the abilities of OpenAI's CLIP neural network. The tool will download a YouTube video, extract frames at regular intervals and precompute the feature vectors of each frame using CLIP. You can then use natural language search queries to find a particular frame of the video. The results are really amazing in my opinion... If you want to experiment with it yourself, I prepared a Colab notebook that can easily be run: https://colab.research.google.com/github/haltakov/natural-language-youtube-search/blob/main/natural-language-youtube-search.ipynb https://colab.research.google.com/github/haltakov/natural-la...

pininja 6y ago
This is very cool! Does this produce a occurrence index by any chance? It would be neat to explore a word map of a video.
- vladoh 6y ago
  Not yet, but I had this idea as well. You basically get a score describing how well a phrase is matching each of the images so it won't be difficult to do. I'll look into that!
- vladoh 6y ago
  I pushed a small update and you will now see a heatmap displaying the score of the search query for each frame.
  - pininja 6y ago
    Looks great! Adds a whole dimension to the video. Thanks.
fireattack 6y ago
Just a heads up: in the demo's setup block, it installs pytorch 1.7.0 + cu101 from CLIP (since it requires them) and then immediately uninstalls it to re-install pytorch 1.7.1 by the next command, which takes at least 5 minutes. If we don't really need 1.7.1, we can save some time by removing the manual pytorch installation line.
- vladoh 6y ago
  Yes, I know that this is a bit slow. The problem is you really need 1.7.1, because 1.7.0 leads to some strange issues and broken results: https://github.com/openai/CLIP/issues/13#issuecomment-771143745 https://github.com/openai/CLIP/issues/13#issuecomment-771143...
  - fireattack 6y ago
    Ah, got it. I just noticed 1.7.1 is already in CLIP's requirements.txt, weird that colab would still install 1.7.0 to begin with. Edit: I just realized 1.7.0 just came with colab, not installed by CLIP.
    - vladoh 6y ago
      Yes, this is the actual problem...
mandeepj 6y ago
> The tool will download a YouTube video, extract frames at regular intervals That should be able to scale well :-)
- woko 6y ago
  For info, the same tool works well with 2 million images found in the Unsplash dataset [1]. Features only have to be computed once for the dataset, and only the feature vector for the user query has to be computed on the fly. Then matching features can be done in a manner that scales well. So, the present tool does not scale because the videos are part of the user query, but a company with an easy access to the videos and the computational power to pre-encode the frames as features could create a search engine based on CLIP. [1] https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search
  - vladoh 6y ago
    Thanks for sharing! :) Yes, the feature computation on the images has to be sone only once and the representation is very efficient - 512 float16 values per image.
    - woko 6y ago
      Yes, I know. :D Your previous project with Unsplash made me try a similar approach [1] for banners of video games on Steam. [1] https://github.com/woctezuma/steam-image-search https://github.com/woctezuma/steam-image-search
      - vladoh 6y ago
        Cool application! I wasn't aware that there are so many images available on Steam...
        woko 6y ago
        Yes, Steam has grown a lot. Last I checked, there were ~50k apps. Edit: 50,630 apps according to https://www.gamedatacrunch.com/ https://www.gamedatacrunch.com/ As I focused on vertical banners, the list was smaller (~30k apps). This is equivalent to the *lite* version of Unsplash's dataset.
mockingbirdy 6y ago
Awesome! I'm currently working on the exact same thing (but with OCR added). Thank you for releasing this.
- canada_dry 6y ago
  > the exact same thing (but with OCR added) Hmmm... what does "with OCR added" mean? If there is text in the video (e.g. street sign) that it can also be searched??
  - mockingbirdy 6y ago
    No, that wouldn't work too well. It's for YouTubers who stream their desktop screens and I need to extract some information to automatically process it. The desktop streams always look very similar so I don't need advanced AI/neural nets to extract that.
canada_dry 6y ago
Great demo. Wondering whether it would be more efficient if extracting frames where the content has changed (e.g. over a threshold and/or all I-frames)? Also, could this be used to identify event type in videos? I'd love to run my 25 years of home videos through this an have it annotate: "Christmas, birthday, park, camping...".
- vladoh 6y ago
  Yes, this is definitely possible. You can maybe try computing some kind of image distance between frames or some keyframe extraction. Once you compute the features, the search is very efficient! I tried it for searching in the 2M photos dataset from Unsplash and it takes like 2-3 seconds: https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search I plan to run my personal photos through it :)
ramraj07 6y ago
This is amazing. I'm going to get this running on my Dropbox. Curious what it gets out of scanned documents as well.
- vladoh 6y ago
  There is one caveat to be aware of - the image is cropped to a square in the center and scaled down to 224x224. So small details will be lost, for example if you want to run it on scanned documents. Photos work great though. I tried it on the 2M photos from the Unsplash dataset: https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search
Crazyontap 6y ago
Can it work for more advance keywords like say, "traffic violation" where it spots a car jumping red light or pedestrian not using a crosswalk, etc? It could be very useful to help with law enforcement.
- vladoh 6y ago
  I think it can. However, you will likely need a bigger model. Currently, OpenAI shares only their small model and I hope they will soon release bigger ones!
Abishek_Muthian 6y ago
Excellent work. If it could take image set as input, then perhaps we can use this to identify our self in a random Internet video e.g. Lengthy tourist video in which you suspect you could have been covered as you were there at that place on that day. There are people already looking for such solution(I've added the link to that discussion on my profile).
- vladoh 6y ago
  I think this is a valuable application, but I don't think CLIP is well suited for it. The power of CLIP comes from training a model to jointly "understand" text and images. If you are looking at identifying a particular person there are more suitable designs for face recognition.
akinhwan 6y ago
Wow what sort of business ideas do you think could come of this?

by vladoh 6y ago

pininja 6y ago

This is very cool! Does this produce a occurrence index by any chance? It would be neat to explore a word map of a video.

vladoh 6y ago
Not yet, but I had this idea as well. You basically get a score describing how well a phrase is matching each of the images so it won't be difficult to do. I'll look into that!
vladoh 6y ago
I pushed a small update and you will now see a heatmap displaying the score of the search query for each frame.
- pininja 6y ago
  Looks great! Adds a whole dimension to the video. Thanks.

fireattack 6y ago

Just a heads up: in the demo's setup block, it installs pytorch 1.7.0 + cu101 from CLIP (since it requires them) and then immediately uninstalls it to re-install pytorch 1.7.1 by the next command, which takes at least 5 minutes. If we don't really need 1.7.1, we can save some time by removing the manual pytorch installation line.

vladoh 6y ago
Yes, I know that this is a bit slow. The problem is you really need 1.7.1, because 1.7.0 leads to some strange issues and broken results: https://github.com/openai/CLIP/issues/13#issuecomment-771143745 https://github.com/openai/CLIP/issues/13#issuecomment-771143...
- fireattack 6y ago
  Ah, got it. I just noticed 1.7.1 is already in CLIP's requirements.txt, weird that colab would still install 1.7.0 to begin with. Edit: I just realized 1.7.0 just came with colab, not installed by CLIP.
  - vladoh 6y ago
    Yes, this is the actual problem...

mandeepj 6y ago

> The tool will download a YouTube video, extract frames at regular intervals That should be able to scale well :-)

woko 6y ago
For info, the same tool works well with 2 million images found in the Unsplash dataset [1]. Features only have to be computed once for the dataset, and only the feature vector for the user query has to be computed on the fly. Then matching features can be done in a manner that scales well. So, the present tool does not scale because the videos are part of the user query, but a company with an easy access to the videos and the computational power to pre-encode the frames as features could create a search engine based on CLIP. [1] https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search
- vladoh 6y ago
  Thanks for sharing! :) Yes, the feature computation on the images has to be sone only once and the representation is very efficient - 512 float16 values per image.
  - woko 6y ago
    Yes, I know. :D Your previous project with Unsplash made me try a similar approach [1] for banners of video games on Steam. [1] https://github.com/woctezuma/steam-image-search https://github.com/woctezuma/steam-image-search
    - vladoh 6y ago
      Cool application! I wasn't aware that there are so many images available on Steam...
      - woko 6y ago
        Yes, Steam has grown a lot. Last I checked, there were ~50k apps. Edit: 50,630 apps according to https://www.gamedatacrunch.com/ https://www.gamedatacrunch.com/ As I focused on vertical banners, the list was smaller (~30k apps). This is equivalent to the *lite* version of Unsplash's dataset.

mockingbirdy 6y ago

Awesome! I'm currently working on the exact same thing (but with OCR added). Thank you for releasing this.

canada_dry 6y ago
> the exact same thing (but with OCR added) Hmmm... what does "with OCR added" mean? If there is text in the video (e.g. street sign) that it can also be searched??
- mockingbirdy 6y ago
  No, that wouldn't work too well. It's for YouTubers who stream their desktop screens and I need to extract some information to automatically process it. The desktop streams always look very similar so I don't need advanced AI/neural nets to extract that.

canada_dry 6y ago

Great demo. Wondering whether it would be more efficient if extracting frames where the content has changed (e.g. over a threshold and/or all I-frames)? Also, could this be used to identify event type in videos? I'd love to run my 25 years of home videos through this an have it annotate: "Christmas, birthday, park, camping...".

vladoh 6y ago
Yes, this is definitely possible. You can maybe try computing some kind of image distance between frames or some keyframe extraction. Once you compute the features, the search is very efficient! I tried it for searching in the 2M photos dataset from Unsplash and it takes like 2-3 seconds: https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search I plan to run my personal photos through it :)

ramraj07 6y ago

This is amazing. I'm going to get this running on my Dropbox. Curious what it gets out of scanned documents as well.

vladoh 6y ago
There is one caveat to be aware of - the image is cropped to a square in the center and scaled down to 224x224. So small details will be lost, for example if you want to run it on scanned documents. Photos work great though. I tried it on the 2M photos from the Unsplash dataset: https://github.com/haltakov/natural-language-image-search https://github.com/haltakov/natural-language-image-search

Crazyontap 6y ago

Can it work for more advance keywords like say, "traffic violation" where it spots a car jumping red light or pedestrian not using a crosswalk, etc? It could be very useful to help with law enforcement.

vladoh 6y ago
I think it can. However, you will likely need a bigger model. Currently, OpenAI shares only their small model and I hope they will soon release bigger ones!

Abishek_Muthian 6y ago

Excellent work. If it could take image set as input, then perhaps we can use this to identify our self in a random Internet video e.g. Lengthy tourist video in which you suspect you could have been covered as you were there at that place on that day. There are people already looking for such solution(I've added the link to that discussion on my profile).

vladoh 6y ago
I think this is a valuable application, but I don't think CLIP is well suited for it. The power of CLIP comes from training a model to jointly "understand" text and images. If you are looking at identifying a particular person there are more suitable designs for face recognition.

akinhwan 6y ago

Wow what sort of business ideas do you think could come of this?