All About Kamua, A New AI-Enabled Video-Editing Platform


Kamua is a new artificial intelligence-powered video-editing platform that is designed to help businesses, marketers, and creators. It helps by democratising the video editing process that requires little technical expertise. With this platform, users can automatically convert landscape-mode videos to vertical suitable for mobile portrait platforms such as Instagram and Snapchat.

Kamua is capable of operating completely on the cloud using fast GPUs. This platform doesn’t require any local computing resources, thereby reducing overheads such as manual workflows and time required. As per the company blog, this platform has helped customers cut manual workflows from up to 16 hours to just 15 minutes through automation.


Sign up for your weekly dose of what's up in emerging technology.

While the product was available as an alpha release since last year, it launched an invite-only beta in September. Now the leadership is looking at an extensive roll-out on December 1 when a limited free version will be released.

How Does Kamua Work?

The platform is built and trained on several neural networks to automate low-added value tasks that video editors may have to perform. Some of these tasks include — finding cut points in videos; cropping videos from landscape format and transforming them into portrait and square footage; and re-arrangement of the order of shots in a video.

Some of its main features include:

  • Automatic cropping: Using a functionality called AutoCrop, Kamua generates multiple aspect ratio outputs from the same focal point. It uses its proprietary AI technique to automatically detect a focal point to be followed.
  • Tracking specific objects: Tool named CustomTrack helps in cropping specific objects from the video clipping. CustomTack also helps in fitting large objects or regions inside a smaller crop box and also helps in refining the output of AutoCrop further.
  • Motion KeyFraming tool: This tool allows users to set coordinates and properties at particular frames and instances in a video clip. It helps the user in moving the crop box across the screen at a controllable pace. Further, this tool also helps in controlling the precise zoom level at each KeyFrame. It can be used along with AutoCrop to alter automatic crops directly.
  • Automatic detection of cut points: Powered by AI, AutoCut helps in automatically decompiling the original video for users to add, remove, or trim clips before finally recompiling. This tool looks for points in the video to detect where the original editor made the alteration.
  • Captions and subtitles: The AI-enabled AutoCaptions tool of Kamua is word-timed which assigns a timestamp to every word in the transcript, and the word appears on the screen as spoken. It also offers users the option to manually insert, split, or correct the subtitles in case of any discrepancies. The team is now working on bringing Netflix style blocks of subtitles.
  • Colour correction: Kamua’s colour correction techniques are currently limited to hue, saturation, brightness, contrast, and tint. Research is on for adding more presets, tuning, rapid colour and correction, and style transfer using AI.

Browser-Based Service

Kamua is an entirely browser-service that is deployed on Google cloud and whose functions such as video processing and AI processing take place on Nvidia GPUs. Its proprietary machine learning algorithms are 95% accurate in addition to the neural networks that are capable of identifying and tracking actions in a given scene.

Kamua uses Google’s Tensorflow, NVIDIA NGX, CUDA, and other open-source computer vision algorithms, tools, and frameworks. In an interview, Kamua CEO and cofounder Paul Robert Cary said that Kamua is also planning offline support in future.

Wrapping Up

Kamua bears great resemblance to Google’s machine learning-based video reframing technology AutoFlip, which the former also noted in a blog. Kamua mentions that Google’s approach to reframing and shot detection is ‘almost exactly’ similar to theirs. However, the Kamua’s team feels that their product is better in the sense that it uses multiple neural networks that collaborate to estimate focal points of the videos and offers better camera smoothing and pathing.

In future, Kamua plans to offer analytics to its customers and introduce a mobile app which will automatically resize videos from the device’s camera roll. Apart from the basic free version that will be available from December, the company will launch a paid version with more cloud storage and other features.

More Great AIM Stories

Shraddha Goled
I am a technology journalist with AIM. I write stories focused on the AI landscape in India and around the world with a special interest in analysing its long term impact on individuals and societies. Reach out to me at

Our Upcoming Events

Conference, in-person (Bangalore)
Machine Learning Developers Summit (MLDS) 2023
19-20th Jan, 2023

Conference, in-person (Bangalore)
Rising 2023 | Women in Tech Conference
16-17th Mar, 2023

Conference, in-person (Bangalore)
Data Engineering Summit (DES) 2023
27-28th Apr, 2023

Conference, in-person (Bangalore)
MachineCon 2023
23rd Jun, 2023

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM