I’m excited to announce that Flame 2021 is here, bringing even more speed and flexibility to compositing, color grading, and finishing workflows. Building on the machine learning-powered features first introduced in Flame 2020, this update adds more uses of AI – this time for isolating features on the human face. By popular demand, we’ve also added a new GPU-accelerated Physical Defocus effect as well as a slew of finishing enhancements to make it easier to adjust looks across many shots, share updates with clients, and create beautiful work, fast. Those of you doing finishing work for streaming services will also now enjoy powerful new workflows for certified Dolby Vision HDR authoring and display.
GPU-Accelerated Defocus Effect
This one was a popular request by many of you! The new Physical Defocus effect makes it easier for you to add depth of field to your shots. It can be used in Batch as a standalone Matchbox node or in Image Selective in combination with a generated AOV Depth map.
You can choose between Physical and Action Camera focus models or create your own custom models by adjusting attributes. Lens artifacting and radial, chromatic aberration can be customized to achieve the level and quality of out-of-focus blur you want, Depth Slices can be increased and blended for more granularity of focus, and highlights can be bloomed using a selection of Bokeh attributes and lens noise models.
Machine Learning-Powered Human Face Segmentation
A new Semantic keyer for feature isolation on the human face adds the ability to extract alpha mattes for skin, eyes, lips, nose, cheeks, chin, and laugh lines.
Useful for compositing, color grading, and cosmetic beauty work, the Human Face Part Extraction Keyer does not require any manual tracking – AI does the heavy lifting for you! In shots with multiple faces, you can use an on-screen widget to guide the AI. All Keyers can be combined and refined with GMasks in Action or the Image node at the timeline level.
For specialized tasks such as isolating a mole or scar, there’s a new custom layout workflow. Using a template for the human face that can be painted on ‘unwrapped,’ you can define custom regions that are then displaced by the face matching algorithm.
Flame 2021 introduces a new way to quickly save and recall color grading and FX work in the Explorer, a dedicated ‘grade bin’ and reference comparison area. Image node Selectives can now be dragged to the Explorer panel or saved by keyboard shortcut. Selective layers can also be saved individually to be recalled further down the timeline or in another sequence entirely.
A new video preview mode called ‘Viewing Area’ makes it easy for you to share iterations of your work to large SDI or HDMI preview monitors. All artist viewports are visible in this mode including storyboard, manager, and schematic. When using Viewing Area in broadcast mode, GMasks can be viewed and edited at the same time, along with any other tool you directly manipulate.
Dolby Vision HDR Authoring and Display Workflow
While there have been ways to output HDR content from Flame using color management and floating-point processing up until this point, this latest update goes a step further by adding seamless workflows for importing, authoring, displaying, and exporting Dolby Vision HDR shot-by-shot animatable metadata.
Check out the new features in Flame 2021 by logging into your Autodesk Account and downloading the latest release. If you are not yet a subscriber, make sure to check out our 30-day free trial of Flame 2021.
Note: It may take up to 3 business days for this release to show up in your Autodesk Account.
Mark Your Calendars: Up Close & Personal with Flame
April 28, 2020 1:00 PM - 3:00 PM ET
Join the Flame team for a first look and discussion around the latest updates to Flame. Special guest Andy Milkis, Director of Visual Effects and Technology at LVLY and co-founder of the New York Flame User Group, will also present the winners of the first-ever LogikFest 2020.
But wait, there's more!
Read our extensive overview of what’s new in Flame 2021.
Visit the Flame Learning Channel for the latest how to’s.
Learn more about how AI is being used for Machine Learning in Flame.