With video becoming the top priority for content investment in 2023, this month’s innovations from Cloudinary focus on empowering brands to scale video experiences to any device or channel. Key features include AI-based cropping to maintain focus on desired objects – especially when viewed on mobile – and normalizing audio quality for video sourced from third parties.
To get more color on these features and to see them in action, visit Cloudinary Academy or our YouTube channel.
For a detailed list of all the latest innovations, check out our release notes.
Close to 75% of visits to e-commerce sites are from Smartphones and 63% of all orders are made on mobile devices (source). Meaning, it is mission-critical that e-commerce brands adapt their product and marketing videos shot in landscape mode to the mobile or portrait view. As anyone who has tried knows, reformatting hundreds or thousands of videos manually is not easy, let alone scalable.
A few years ago Cloudinary launched AI-based video cropping with auto-gravity to address the reformatting issue. This advanced AI feature identifies the most interesting areas in a video then stays focused on that item the duration of the video, adjusting the size or aspect ratio to fit any mobile device.
The latest feature, AI Video Cropping: Auto-gravity for objects, handles more complex scenarios in which there may be multiple interesting objects in the video. By using the parameter <g_auto:object> in the video URL, the cropping algorithm can be adjusted to focus on a specific object rather than the most interesting area in the video.
Compare the two cropped versions of the video below. In the video on the left, using g_auto crops the video to maintain focus on the player receiving and kicking the ball. In the video on the right, specifying g_auto:ball ensures the focus is maintained on the ball while cropping the video.
This feature offers several benefits:
- Reduces the amount of time and effort spent manual cropping while maintaining control around the cropping process
- Enables the creation of multiple variants from a single video highlighting different products in each variant
- Improves visual quality of the video by capturing fine movements with sub-pixel level accuracy
- Reduces the file size of the video, making it easier to upload and share
Watch this YouTube video to see this feature in action!
Whether it’s purchasing an electronic appliance or a complex financial service, most people (94%) watch explainer videos and 84% of the viewers decide to purchase after watching the video (source: Hubspot report). Brands often source these explainer videos from experts reviewing the product or from loyal users. Because explainer videos are often user-generated, visual and audio quality is not always that good.
Cloudinary offers on-the-fly adaptive bitrate streaming (ABR) to ensure that the video quality and format adapts in real time for the best visual experience and flawless streaming. Additionally, Cloudinary now offers audio normalization with on-the-fly ABR to ensure smooth transitions between varying audio levels and deliver audio at a uniform volume in user-generated videos.
What once required expensive video editing software and expertise, Cloudinary audio normalization leverages AI and programmatic URL-based transformation to modulate volume. Brands can perform the complex task of audio normalization automatically and at scale. All that’s needed is to add the <e_volume:auto> parameter to the video URL to ensure consistent volume levels throughout the video.
Watch this YouTube video to see this feature in action!
That wraps up the key video innovations for March. Stay tuned for more updates! If you are new to Cloudinary and want to learn more, sign up for a free account.
