Help Center

Welcome to Everypixel Labs

Everypixel Labs offers a range of advanced AI-powered services. Our services are built around a simple idea: you provide input data (an image, video, audio, or text) to a service and it returns predictions or modifies the data.

You can learn more about our algorithms and get instructions for seamless integration with our services in our API documentation. If you have any questions or need any assistance, please contact us at api@everypixel.com.

Getting Started

Before you start working with Everypixel Labs, you need to register an account. Then you can get your Client ID and Secret keys by logging into your Everypixel Labs account and going to API Keys.

Your Client ID and Secret authenticate your API requests so that other people cannot use them. Remember that your Client ID and Secret must be kept secret at all times.

You can also use OAuth 2.0 authentication to increase security.

Billing

After registering your account, you will have free limits to test our algorithm. Each algorithm has its own limits.

Limits will not be renewed once used. This applies to both API access and the use of algorithms through the Everypixel Labs interface. Once the free limit is exceeded, any additional usage will require payment. To continue using Everypixel Labs algorithms, simply top up your balance through the Balance menu in your account by clicking the Top up balance button.

If your balance is low, we will send a warning email. You can determine the minimum balance limit in your profile on the Balance page to let us know when to send you a warning.

After exceeding the number of available requests from your plan, you will see error 429.

Auto-payment

To make sure your balance is always topped up, you may want to consider enabling Auto-payment.

Why enable auto-payment

1. Automatically top up your balance without manual intervention.
2. Set threshold, one-time payment amount, and monthly limits to suit your needs.
3. As with manual balance top-ups, your payment details are kept secure, and you retain full control over your payments.

How to set up auto-payment

Before setting up, make sure you have a Payment method and Billing info on file. To fill in these details, go to the right menu and select Billing.

1. Navigate to the Balance menu in your account.
2. Enter the following details in the Auto-payment block:
   Replenishment threshold: The balance amount at which auto-payment should be triggered.
   One-time payment: The amount to be automatically debited when the threshold is reached.
   Monthly limits: The maximum amount of auto-payments allowed per calendar month.
3. Click on the Save and activate button. An email notification will confirm that Auto-payment has been activated.

How to turn auto-payment off

1. Navigate to the Balance menu in your account.
2. In the Auto-payment block click the Disable button. The button will revert to Save and activate, indicating that Auto-payment has been disabled.

How to update auto-payment details

1. Navigate to the Balance menu in your account.
2. In the Auto-payment block click the Disable button. The button will revert to Save and activate, indicating that Auto-payment has been disabled. This allows you to update your auto-payment details and then re-enable the option.
3. Adjust the replenishment threshold, one-time payment and monthly limits as needed.
4. Click Save and activate to apply the changes. You will receive an email notification confirming that Auto-payment has been activated.

Steps to take if an auto-payment fails

1. Review your billing history in the Billing menu and auto-payment details in the Balance menu. Ensure your monthly limit has not been exceeded.
2. Verify your payment method in the Billing menu. Make sure your payment method is on file, up to date and has sufficient funds.
3. Check your billing info in the Billing menu. Ensure that your billing address is on file and current.

API Request Requirements

If you are using the GET method to submit an image, you must first encode its URL. We recommend using the Urlencode function to separate the image URL from the request.

This is required if the image URL contains special characters such as '&' or '? For example, if your image has the following address

https://images.unsplash.com/photo-1564219563357-e29c471253b6?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=MnwxMTI4NzN8MHwxfGFsbHwyMjI3Nzh8fHx8fHwyfHwxNjgzMTg1MDY3&ixlib=rb-4.0.3&q=80&w=400.

This URL contains special characters '&' or '?'. And if you pass it as a parameter to

https://api.everypixel.com/v1/keywords?url=<url>, you’ll get an error.

The algorithm can process the image if the URL is in URL-encoded form:

https%3A%2F%http://2fimages.unsplash.com/%2Fphoto-1564219563357-e29c471253b6%3Fcrop%3Dentropy%26cs%3Dtinysrgb%26fit%3Dmax%26fm%3Djpg%26ixid%3DMnwxMTI4NzN8MHwxfGFsbHwyMjI3Nzh8fHx8fHwyfHwxNjgzMTg1MDY3%26ixlib%3Drb-4.0.3%26q%3D80%26w%3D400.

When using the POST method, URL encoding is not required.

Privacy and Responsible Use of Technology

We maintain the highest security standards to protect your information. We don't store your images and files when you submit an API request.

At Everypixel, we are also committed to providing AI algorithms that empower professionals and make their work more effective and easier. We do not allow our AI technology to be used for criminal or unethical purposes.

Image Keywording

AI to recognize objects, people, places and actions in images and turn them into keywords.

Pricing

Processing Speed

You can submit simultaneously no more than 30 requests per second. If you exceed this limit, you will receive a 502 error. The maximum number of words the algorithm can generate is 50.

The size of the images you submit will affect the speed of the API. Therefore, we recommend pre-compressing images to 300px on the short side.

Image Requirements

Formats: .png and .jpg.

When processing an image, the algorithm compresses it to 300x300px. This resolution allows us to keep recognition accuracy and API speed high. If you upload images in a higher resolution, it will not affect the number or relevance of keywords, but it will slow down the API speed.

The algorithm is trained on images with real objects or those that have the outlines of real objects. As a result, it may be poor at recognizing objects that do not exist in our world. For example, fantasy-style images: the algorithm may not recognize some magical creatures like a unicorn or a dragon, and will most likely describe the image only with general keywords like "illustration", "fantastic plot", and so on. But the algorithm will successfully recognize illustrations if the details on them have the outlines of real objects.

Variable Request Parameters

num_keywords — you can specify the maximum number of words to be returned. The maximum number of words the algorithm can produce is 50. At the same time, the algorithm independently determines the exact number of keywords to return.

threshold — you can set a minimum value of keyword relevance (score) below which the algorithm will not return keywords.

If you set both parameters at the same time, e.g. threshold=0.5&num_keywords=10, then the algorithm will either generate 10 keywords, or, if the specified score is 0.5, it will generate 7 keywords with a score higher than 0.5.

Getting Results

keyword — the keyword that the algorithm has determined for your image.

score — the degree of relevance of the keyword to your image. The higher the score, the more precisely the keyword is defined and the higher it is on the list.

Video Keywording

AI to recognize objects, people, places and actions in video and turn them into keywords.

Pricing

Video Requirements

Formats: .mp4, .mpeg, .mov, and .avi up to 100 Mb.

The method successfully works only with video files with a duration of 5 to 60 seconds. Increasing the duration of the video decreases the relevance of keywords.

Variable Request Parameters

num_keywords — you can determine the maximum number of words to return. The maximum number of words that the algorithm can produce is 50. At the same time, the algorithm independently determines the exact number of keywords to be returned.

threshold — you can determine a minimum keyword relevance value (score), below which the algorithm will not return keywords

If you determine both parameters at the same time, for example, threshold=0.5&num_keywords=10, then the algorithm will produce either 10 keywords, or, if the determined score 0.5 occurs earlier, it will produce 7 keywords with a score higher than 0.5.

Getting Results

keyword — the keyword that the algorithm has determined for your video.

score — the degree of relevance of the keyword to your video. The higher the score, the more precisely the keyword is defined and the higher it is on the list.

Stock Photography Scoring

This model evaluates the quality and market potential of professional photographs from the perspective of stock photography experts. It does not evaluate how attractive a person or object looks in the photo, but focuses solely on technical aspects such as brightness, contrast, and noise. The model is not intended to evaluate historical photos, illustrations or 3D visualizations.

Pricing

Processing Speed

You can submit simultaneously no more than 30 requests per second. If you exceed this limit, you will see error 502.

Image Requirements

Formats: .png and .jpg.

Getting Results

score — this parameter indicates how well the photo meets the quality, relevance, aesthetic and style criteria established by the designers for stock photos. The score is expressed as a decimal value, with 1.0 representing 100%. A high score indicates that the photo meets the desired standards, while a lower score may reflect technical perfection but indicate problems such as being overly staged or outdated, which would prevent it from achieving a high rating.

UGC Photography Scoring

The main difference between Stock Photo Scoring and this model is the training dataset. User-generated photo scoring is a model trained on 347,000 user photos from Instagram. The estimation parameters for this model were created by a group of 10 professional photographers.

This model is designed to score user photos taken with both a professional camera and a smartphone camera. It doesn't evaluate the action and doesn't measure how cool or beautiful a person or object looks in a photo. It only cares about technical parts like brightness, contrast, noise, and so on. The service is not designed to rate historical photos, illustrations, or 3D visualizations.

Pricing

Processing Speed

You can submit simultaneously no more than 30 requests per second. If you exceed this limit, you will see error 502.

Image Requirements

Formats: .png and .jpg.

Getting Results

score — this parameter indicates the evaluation of the quality of the user's photo based on technical factors such as brightness, contrast and noise. The score is expressed as a decimal value between 0 and 1, where 1.0 is a perfect score (100%).

class — this parameter categorizes the photo into one of five quality classes based on the score and can be expressed as a number from 1 to 5, corresponding to:

1. Very bad (0-20%): Indicates poor quality with significant technical problems.
2. Poor (20-40%): Indicates below average quality with noticeable flaws.
3. Normal (40-60%): Reflects acceptable quality, meeting basic standards but lacking distinction.
4. Good (60-80%): Represents high quality with minor technical imperfections.
5. Excellent (80-100%): Represents the highest quality, with minimal defects and strong technical performance.

Age Recognition

This model understands how old a person looks. It extracts facial features from a photo and provides an accurate estimate of age. You also get a confidence score for facial recognition.

Pricing

Processing Speed

You can submit simultaneously no more than 30 requests per second. If you exceed this limit, you will see error 502.

Image Requirements

Formats: .png and .jpg.

Text-to-Speech

This model makes it easy to convert text to speech and integrate it into business, sales, or customer support processes, as well as applications in a variety of technical environments.

Pricing

Processing Speed

The algorithm will process 1 minute of content in about 10 minutes. It's important to note that processing is sequential, so if you upload multiple videos at once, they will be queued and processed one at a time.

Lipsync

By merging a video source and voiceover, this model accurately reproduces a person's lip movements in a video with speech in multiple languages.

Pricing

Processing Speed

The algorithm will process 1 minute of content in about 10 minutes. It's important to note that processing is done sequentially, so if you upload multiple videos at once, they will be queued and processed one at a time.

File Requirements

Audio:

• Formats: .mp3 and .wav.
• No music: Ensure no music is included.
• Minimal background noise: Aim for clear, noise-free recordings.
• Clear voice recordings: The clearer the voice, the better the LipSync quality.

Video:

• Formats: .mp4 or .mov.
• Resolution: Select videos in HD or higher.
• Stability: Opt for videos with minimal camera movement.
• Head movement: Prefer videos where the head is mostly stationary, with pans less than 40 degrees.
• Mouth visibility: Ensure the mouth isn't covered for long periods of time in the video.
• Lighting: Pick videos where the face is evenly and fully lit.
• Eye contact: Videos with minimal eye squinting are ideal.

Image Captioning

The model generates bulk image descriptions tailored for blog posts, social media, and marketplaces.

Pricing

Processing Speed

You can submit simultaneously no more than 30 requests per second.

Image Requirements

Formats: .png and .jpg.

Image Face Swap

This algorithm transfers a face from one photo to another, aligning facial features to create a realistic result.

Pricing

Face Swap is the process of simulating a face, rather than simply transferring it. Neural networks analyze two subjects:

Donor — the person whose face is to be generated.

Double — the person whose face is used as the basis for the generation.

Image Requirements

Formats: .png and .jpg.

Technology Limitations

Head shape: For best results, the double should have a similar head shape to the donor. Consider factors such as cheekbones, chin, brow ridges, and forehead height.

Hairstyle: Hair and hairstyles are not transferred.

Close-ups: In close-ups (where the face takes up more than 1/6 of the frame), more time is needed to detail features such as wrinkles and facial expressions. If the face is too small, it is harder for the neural network to recognize the contour of the face.

Objects and accessories: The face should not be obstructed by objects. Glasses, beards, and other accessories can affect the results.

Color Matching: The skin tones of the donor and the double should match closely to avoid unnatural color differences in the final result.

Video Face Swap

This algorithm transfers a face from one photo to another, aligning facial features to create a realistic result.

Pricing

Face Swap is the process of simulating a face, rather than simply transferring it. Neural networks analyze two subjects:

Donor — the person whose face is to be generated.

Double — the person whose face is used as the basis for the generation.

File Requirements

Image:

• Formats: .png and .jpg.

Video:

• Formats: .mp4, .mpeg, .mov, and .avi.
• Duration: About 3 minutes, capturing a full range of emotions and head turns.
• Quality: Avoid sudden movements or blurry frames.
• Emotions and angles: Each emotion should be captured from multiple angles, including neutral shots (e.g., during a conversation).

Technology Limitations

Head shape: For best results, the double should have a similar head shape to the donor. Consider factors such as cheekbones, chin, brow ridges, and forehead height.

Hairstyle: Hair and hairstyles are not transplanted.

Head turning: If the double turns his or her head too much, the neural network may not be able to recognize the face and the image may not be processed.

Close-ups: In close-up shots (where the face takes up more than 1/6 of the frame), more time is needed to detail features such as wrinkles and facial expressions. If the face is too small, it is harder for the neural network to recognize the face's contour.

Video resolution: At very high resolutions (such as 4K), the face may appear blurred due to technical limitations.

Objects and accessories: The face should not be obstructed by objects. Glasses, beards, and other accessories can affect the results.

Color matching: The skin tones of the donor and the double should match closely to avoid unnatural color differences in the final result.

Errors

Most of the errors returned by the API contain text in json format.

In addition, you may receive HTTP status codes with the following errors:

400: This error occurs when the request is incorrect, such as including the wrong type of argument.

401: This error occurs if the login credentials are incorrect or the access token has expired.

403: This error occurs if the account is not activated or access to the API endpoint is restricted for some reason.

415: This error occurs if the file cannot be opened due to being broken or in an unsupported format.

429: This error occurs if you have exceeded the number of available requests in your plan.

500: This error occurs due to an issue on our side. Please try again later.

502: This error occurs if you have reached the maximum number of concurrent requests or if the API is not functioning properly.

503: This error occurs if the service is temporarily unavailable.

If you encounter an error not listed here, please contact our support team via email at api@everypixel.com for assistance.