This add-on is operated by a third party: {{provider}}, and Cloudinary is not responsible for it in any way.
By subscribing to this add-on, you accept {{provider}}'s terms of service and agree that Cloudinary shall be not responsible or liable in respect to this add-on.

Take advantage of Cloudinary's sophisticated deep learning algorithms to accurately and precisely remove the background of almost any image in seconds, leaving you with just the main object in the form of a transparent PNG image.

Activate the add-on either by setting the background_removal parameter to cloudinary_ai in your upload or update method, or by setting the effect parameter to background_removal in your delivery URL.

For example, removing the background on this unicorn photo via the upload method, storing only the background-removed image:

Copy to clipboard
Cloudinary::Uploader.upload("cld_unicorn.jpg", :use_filename => "true", :background_removal => "cloudinary_ai")

Or, removing the background from the original image in Cloudinary on the fly:

Read more...

Take advantage of Cloudinary’s powerful AI-based object detection and content-aware algorithms to enable features such as object-aware cropping and automatic image tagging. With the ability to select specific object models and even versions of the models, the Cloudinary AI Content Analysis add-on provides the flexibility needed to interact with the content that is important to you.

Example: Use the Ruby SDK to automatically apply tags to an image on upload, based on any objects or categories detected by version 1 of the COCO model, above a confidence level of 0.6:

Copy to clipboard
Cloudinary::Uploader.upload("bike.jpg", :detection=>”coco_v1”, :auto_tagging => 0.6)

The result includes all automatically detected objects and the assigned tags above the confidence level:

Copy to clipboard
“tags”: [“backpack”, “bicycle”, “person”],
“data”: {"coco": {"model_name": "coco", "model_version": 1, "schema_version": 1, "tags": {"backpack": [{"bounding-box": [1700.7, 530.5, 177.6, 295.6], "categories": ["accessory"], "confidence": 0.742 }], "bicycle": [{"bounding-box": [213.0, 760.3, 864.8, 514.4], "categories": ["vehicle"], "confidence": 0.997 }], "person": [{"bounding-box": [1304.0, 395.6, 566.4, 589.6], "categories": ["person"], "confidence": 0.998 }]}}}

Example: Use version 1 of the COCO model for object-aware cropping to deliver a square image that keeps the bike in the image, using the Ruby SDK:

Copy to clipboard
cl_image_tag("docs/bike.jpg", :width=>1000, :aspect_ratio=>"1", :gravity=>"coco_v1_bicycle", :crop=>"crop")
Original
g_auto:classic
g_auto:bicycle
Read more...

Optimize your media management with AI Vision by Cloudinary, a powerful tool that automates the analysis, tagging, and moderation of visual content.

Utilizing advanced AI technology, AI Vision offers scalable solutions for handling large volumes of media assets. Whether you need to classify images with custom tags, automate content moderation for compliance, or extract in-depth insights from visuals, AI Vision delivers the flexibility and precision required to meet your business needs, all within a seamless workflow.

Example: Tagging
Use AI Vision to automatically tag images based on provided definitions:

Copy to clipboard
POST https://api.cloudinary.com/v2/analysis/{cloud_name}/analyze/ai_vision_tagging
{
  "source": {
    "uri": "https://res.cloudinary.com/demo/image/upload/sample.jpg"
  },
  "tag_definitions": [
    {"name": "price-tag", "description": "Does the product have visible price tags or labels in the image?"},
    {"name": "accessories", "description": "Does the image include any fashion accessories, such as bags, belts, or jewelry?"},
    {"name": "side", "description": "Is the person side-facing, showing the profile view?"}
  ]
}

Example: Moderation
Evaluate images against specific moderation questions:

Copy to clipboard
POST https://api.cloudinary.com/v2/analysis/{cloud_name}/analyze/ai_vision_moderation
{
  "source": {
    "uri": "https://res.cloudinary.com/demo/image/upload/sample.jpg"
  },
  "rejection_questions": [
    "Does this image contain violence?",
    "Does this image contain nudity?"
  ]
}

Example: General
Gain insights from images by asking open-ended questions:

Copy to clipboard
POST https://api.cloudinary.com/v2/analysis/{cloud_name}/analyze/ai_vision_general
{
  "source": {
    "uri": "https://res.cloudinary.com/demo/image/upload/sample.jpg"
  },
  "prompts": [
    "Describe the image in detail",
    "How many flowers are in the image?"
  ]
}
Read more...

Automatically categorize and tag videos and/or generate transcripts and subtitles with the Microsoft Azure Video Indexer add-on.

* Scenes are categorized, with tags automatically added to the video.
* Audio is transcripted and subtitles automatically uploaded as a separate file

For example, uploading a video for categorization, auto-tagging, and transcription:

Copy to clipboard
Cloudinary::Uploader.upload("dog.mp4", :resource_type => :video, :categorization => "azure_video_indexer", :auto_tagging => 0.7, :raw_convert => "azure_video_indexer")

Microsoft Automatic Video

The result includes the automatic categorization and assigned tags:

Copy to clipboard
"tags": [ "dog", "animal" ...],
"labels": [{"name": "dog", "language": "en-US", "instances": [{"confidence": 0.9556, "start": "0:00:02.135", "end": "0:01:25.419", "duration": "0:01:23.284"}]}]

1 unit = up to 1 minute of audio analysis.
3 units = up to 1 minute of video analysis, excluding audio.
4 units = up to 1 minute of video analysis, including audio.

Note that partial minutes are rounded up.

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically moderate all videos uploaded to your application using Google's AI Moderation. Protect your users from explicit and suggestive adult content in images on your website or mobile application. Use Cloudinary's upload and delivery APIs knowing that Google's algorithms have approved your videos for user viewing.

For example, uploading a video that should be moderated by Google:

Copy to clipboard
  Cloudinary::Uploader.upload("user_video.mp4", resource_type: :video, :moderation => 'google_video_moderation', :notification_url => 'https://mysite/mycallback')
Google AI video moderation approved
Google AI video moderation rejected

* Note: 1 unit = up to 1 minute (partial minutes are rounded up).

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically categorize and tag images using the Google Auto Tagging add-on. Scene categories are identified and assigned to uploaded images, and then Cloudinary automatically adds tags based on these categories.

For example, uploading an image with categorization and automatic tagging:

Copy to clipboard
Cloudinary::Uploader.upload("ice_skating.jpg", :categorization => "google_tagging", :auto_tagging => 0.6)
Google Automatic Tagging

The result includes all automatically detected categories and the assigned tags of high confidence level:

Copy to clipboard
"tags": [ "skating", "footwear", "ice skating" … ],
"data": [ { "tag": "skating", "confidence": 0.964 },
          { "tag": "footwear", "confidence": 0.956 },
          { "tag": "ice skating", "confidence": 0.943 }, ...

Read more...
Select Add-on Plan:
Larger plans? Contact us

Generate automatic transcripts and subtitles for your videos with the help of the neural network-based Google Cloud Speech API.

Generate the transcript during an upload or update call.
For example:

Copy to clipboard
Cloudinary::Uploader.upload("imagecon_grigsby_intro.mp4", :resource_type => :video, :raw_convert  => "google_speech")

Then include the automatically generated subtitles when you deliver your video:

Copy to clipboard
http://res.cloudinary.com/demo/video/upload/l_subtitles:imagecon_grigsby_intro.transcript/imagecon_grigsby_intro.mp4

* Note: 1 unit = up to 15 seconds (partial intervals are rounded up).

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically categorize and tag videos using the Google Automatic Video Tagging add-on. Scene categories are identified and assigned to uploaded videos, and then Cloudinary automatically adds tags based on these categories.

For example, uploading a video for categorization and auto-tagging:

Copy to clipboard
Cloudinary::Uploader.upload("turtle.mp4", :resource_type => :video, :categorization => "google_video_tagging", :auto_tagging => 0.7)
Google Automatic Video Tagging

The result includes all automatically detected categories and assigned tags of high confidence level:

Copy to clipboard
"tags": [ "turtle", "animal" ...],
"data": [
  {"tag": "turtle", "start_time_offset"=>0.0, "end_time_offset"=>13.44,
    "confidence": 0.93},
  {"tag": "animal", "start_time_offset"=>0.0, "end_time_offset"=>13.44,
    "confidence": 0.93} ...  ]

* Note: 1 unit = up to 60 seconds (partial minutes are rounded up).

Read more...
Select Add-on Plan:
Larger plans? Contact us

Use the Google Translation add-on in conjunction with any of Cloudinary's automatic tag generation add-ons to translate the detected tags into any supported language or into multiple languages. Translations are generated using the Google Cloud state-of-the-art neural machine translation. For example, uploading an image and requesting English, French, and Spanish translations of the tags detected by the Google Auto Tagging add-on:

Copy to clipboard
  Cloudinary::Uploader.upload("windmill_day.jpg",
    :categorization => "google_tagging:en:fr:es", :auto_tagging => 0.6)

The results include tags that exceed the requested confidence, translated into the requested languages, as well as translations for each identified category at all levels of confidence. For example:

Copy to clipboard
"tags": [ "windmill", "moulin à vent", "molino",
          "voie navigable",  "camino acuático", "grassland",
          "prairie", "pradera", "sky", .... ],
"data": [   {
              "tag": { "en": "windmill",  "fr": "moulin à vent", "es": "molino" },
              "confidence": 0.9753, ... }]

* Note: 1 unit = 1 translated character.

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically moderate all photos uploaded to your application using Amazon Rekognition's Deep Learning algorithms. Protect your users from explicit and suggestive adult content in images on your website or mobile application. Use Cloudinary's upload and delivery APIs knowing that Amazon Rekognition's Artificial Intelligence algorithms have approved your images for user viewing.

For example, uploading an image that should be moderated by Amazon Rekognition:

Copy to clipboard
Cloudinary::Uploader.upload("user_image.jpg",
:moderation => 'aws_rek', :notification_url => 'https://mysite/mycallback')

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically moderate all videos uploaded to your application using Amazon Rekognition's Deep Learning algorithms. Protect your users from explicit and suggestive adult content in videos on your website or mobile application. Use Cloudinary's upload and delivery APIs knowing that Amazon Rekognition's Artificial Intelligence algorithms have approved your videos for user viewing.

For example, uploading a video that should be moderated by Amazon Rekognition:

Copy to clipboard
Cloudinary::Uploader.upload("user_video.mp4", resource_type: :video,
:moderation => 'aws_rek_video', :notification_url => 'https://mysite/mycallback')

* Note: Partial seconds are rounded up.

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically categorize and tag images using the Amazon Rekognition Auto Tagging add-on. Scene categories are identified and assigned to uploaded images, and then Cloudinary automatically adds tags based on these categories.

For example, uploading an image with categorization and automatic tagging:

Copy to clipboard
Cloudinary::Uploader.upload("koala5.jpg",
  :categorization => "aws_rek_tagging", :auto_tagging => 0.7)
Amazon Rekognition Auto Tagging

The result includes all automatically detected categories and assigned tags of high confidence level:

Copy to clipboard
"tags": [ "Animal", "Koala", "Mammal" ], ...
"data": [ { "tag": "Animal", "confidence": 0.9933 },
          { "tag": "Koala", "confidence": 0.9933 },
          { "tag": "Mammal", "confidence": 0.99333 }, ...

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically recognize and tag images using the Amazon Rekognition Celebrity Detection add-on. Celebrities are identified in your uploaded images, and then Cloudinary automatically tags the images with the celebrity names.

For example, uploading an image with celebrity detection and automatic tagging:

Copy to clipboard
Cloudinary::Uploader.upload("golfer.jpg",
  :detection => "aws_rek_face",
  :auto_tagging => 0.8)
Amazon Rekognition Celebrity and Face Detection

The result includes all automatically detected celebrities and assigned tags meeting the confidence level:

Copy to clipboard
"tags": [ "Tiger Woods" ], ...
"data": [ { "name": "Tiger Woods", "match_confidence": 100 }, ...

Read more...
Select Add-on Plan:
Larger plans? Contact us

Apply transformations to text detected in your images using the OCR Text Detection and Extraction add-on (powered by the Google Vision API). Pixelate, blur or cover phone numbers, email addresses, license plates, and other undesired data. And of-course you can extract the detected text content and coordinates from any image, including multi-page documents like TIFFs and PDFs.

For example, pixelating the numbers and letters on a license plate:

Copy to clipboard
cl_image_tag("black_car.jpg", :effect=>"pixelate_region:15", :gravity=>"ocr_text")
Original
OCR original image
Pixelated
OCR Pixelated image

Same example as a dynamic URL:

Read more...

Imagga's smart cropping technology automatically chooses the most visually appealing parts of your images and crop them accordingly. A great addition to Cloudinary's face detection based cropping. Any Cloudinary's image manipulation can be applied on the cropped images.

For example, embedding a 260x240 smartly cropped image:

Copy to clipboard
cl_image_tag("sample_van.jpg", :crop => "imagga_scale",
             :width => 260, :height => 240)
Original
Imagga original image
Cropped
Imagga cropped image

Same example as a dynamic URL:

Read more...
Select Add-on Plan:
Larger plans? Contact us

Capture screenshots of any public web page using URL2PNG. Further manipulate your screenshots with any of Cloudinary's image transformations to perfectly match your website's graphic design. Embed on-the-fly generated images in your site, delivered through a fast CDN.

For example, embedding a 300x225 screenshot image:

Copy to clipboard
cl_image_tag("http://cnn.com/", :type => 'url2png',
 :width => 300, :height => 225, :crop => :fill, :radius => 10)
URL2PNG sample screenshot

Same example as a dynamic URL:

Read more...

Automatically moderate all photos uploaded to your application using WebPurify. Prevent adult-oriented and inappropriate images from creeping into your websites or mobile applications. Use Cloudinary's upload and delivery APIs knowing that human moderation experts approved your images for user viewing.

For example, uploading an image that should be moderated by WebPurify:

Copy to clipboard
Cloudinary::Uploader.upload("user_image.jpg",
  :moderation => 'webpurify', :notification_url => 'https://mysite/mycallback')
WebPurify image moderation
Read more...

Automatically detect eyes, nose and other facial attributes in your photos using Advanced Facial Attribute Detection utilizing Microsoft's Cognitive Services. Detect gender, age and other semantic face details and combine them with Cloudinary's image manipulation capabilities to smartly crop images and place overlays over detected faces and eyes.

For example, adding overlays of glasses, automatically positioned, scaled and rotated:

Copy to clipboard
cl_image_tag("coupled.jpg", :overlay => "glasses",
  :width => 1.7, :flags => 'region_relative',
  :gravity => 'adv_eyes')
Original
Advanced facial attributes detection original image
Eyes overlay
Advanced facial attributes detection with eyes overlay

Same example as a dynamic URL:

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically convert Word, Excel and PowerPoint Office documents to PDF or any image format using Aspose. Generated images can be further processed by Cloudinary to create thumbnails of any page in the document. Manipulate your thumbnails to perfectly match your graphic design.

For example, uploading a Word document and converting to it PDF:

Copy to clipboard
Cloudinary::Uploader.upload("my_file_name.docx",
  :public_id => "sample_document.docx",
  :resource_type => 'raw', :raw_convert => 'aspose')
Aspose Office document thumbnail

A JPG thumbnail of the document can now be generated and delivered via a CDN:

Read more...
Select Add-on Plan:
Larger plans? Contact us

Use the Perception Point add-on to scan images, videos, documents, and other files uploaded to your account, protecting you against viruses, spyware, malware, and other malicious code.

The Perception Point add-on leverages multiple antivirus engines simultaneously
to prevent viruses and malware from infecting your website
or mobile application. You can then deliver your files knowing that they
are approved for user viewing and downloading.

For example, uploading a PDF document that will also be scanned
by Perception Point:

Copy to clipboard
Cloudinary::Uploader.upload("user_document.pdf",
  :moderation => 'perception_point',
  :notification_url => 'https://mysite/mycallback')
Metascan antivirus
Read more...

Select Add-on Plan:
Larger plans? Contact us
*The 'Free' plan for this add-on is available only to customers with a paid Cloudinary plan.

Remove the background from your photos, focus only on their main subjects and improve your conversion rates. Remove-The-Background employs expert graphic designers to quickly edit your photos to perfection. The original images will be automatically replaced with their background removed versions as soon as their editing is finished.

Cloudinary will handle all the additional editing, optimizations and delivery of the resulting images.

For example, uploading an image to the cloud while scheduling background removal:

Copy to clipboard
Cloudinary::Uploader.upload("wood_chair.jpg",
  :background_removal => 'pixelz')
Original
Remove The Background original image
Edited
Remove The Background edited image
Read more...
Select Add-on Plan:
Larger plans? Contact us

With VIESUS™ image enhancement add-on, you can extend Cloudinary's powerful image manipulation and optimization capabilities by enhancing images to their best visual quality and automatically adjusting brightness and color, restoring sharpness, removing noise and correcting for overexposure or underexposure.

For example, embedding a 300px wide visually enhanced image:

Copy to clipboard
cl_image_tag("beach.jpg", :effect => "viesus_correct",
             :width => 300, :crop => :scale)
Original
VIESUS original image
Enhanced
VIESUS enhanced image

Same example as a dynamic URL:

Read more...
Select Add-on Plan:
Larger plans? Contact us

Automatically categorize and tag images using the Imagga Auto Tagging add-on. Scene categories are identified and assigned to uploaded images, and then Cloudinary automatically adds tags based on these categories.

For example, uploading an image with categorization and automatic tagging:

Copy to clipboard
Cloudinary::Uploader.upload("turtles.jpg",
  :categorization => "imagga_tagging",
  :auto_tagging => 0.4)
Imagga Auto Tagging

The result includes all automatically detected categories and the assigned tags of high confidence level:

Copy to clipboard
"tags": [ "sea turtle", "loggerhead", "turtle" ], ...
"data": [ { "tag": "sea turtle", "confidence": 1.0 },
          { "tag": "loggerhead", "confidence": 1.0 },
          { "tag": "turtle", "confidence": 0.8903 }, ...

Read more...