Hello... fellow Mavic Mini pilot here.
Grading my footage I've quickly found out that... while I'd truly like to apply LUTs to my videos, I probably couldn't find anything universal that could suit my mood, my scenes, lights and so on. So I resort to manual everything instead.
As for the software, I use Sony Movie Studio Platinum 13 (from 2013)... I don't plan a paying upgrade right now and as much as I like DaVinci Resolve to color grade my footage I am not comfortable with it, I have my small share of knowledge with MS acquired in years and I don't want to give that up and spend months learning again from the basics.
So here I am, sharing my stack to improve my drone videos with a focus on Cinema like pictures. Please consider that I've barely ever checked a tutorial and learnt most of it the hard way.
So I start with a straight out of the camera (Mavic Mini) motion picture still frame (grab), that is my base.
After the basic RAW footage has been trimmed to length there are several passages:
Don't tell me they're in the wrong order, but they do apply quite good in my workflow, and if I try shuffling them around I end up with unsatisfactory results.
Then... I make a copy of it all, and use it as a base for the Sony Gaussian blur filter. That (source alpha, compositing child) track layer is applied selectively with an upper track layer (multiply mask, compositing parent) with a Sony Color gradient mask applied with Fx Mask generator (Alpha). This allows me to simulate the depth of field of a 35mm Cinema camera with Guassian filter of just 0.0015 to 0.0030 range (both horizontal and vertical).
The uppermost layer is just a regular alpha channel with a Color gradient mask to better grade wider regions (e.g. skies, dark expanses).
So far this worked ok for me and I am decently pleased with the result. I could surely improve, save for the fact that I am using outdated software, but I am far more interested in making videos rather than learning new techniques right now. I hope you can enjoy this insight into a stranger's workflow.
Grading my footage I've quickly found out that... while I'd truly like to apply LUTs to my videos, I probably couldn't find anything universal that could suit my mood, my scenes, lights and so on. So I resort to manual everything instead.
As for the software, I use Sony Movie Studio Platinum 13 (from 2013)... I don't plan a paying upgrade right now and as much as I like DaVinci Resolve to color grade my footage I am not comfortable with it, I have my small share of knowledge with MS acquired in years and I don't want to give that up and spend months learning again from the basics.
So here I am, sharing my stack to improve my drone videos with a focus on Cinema like pictures. Please consider that I've barely ever checked a tutorial and learnt most of it the hard way.
So I start with a straight out of the camera (Mavic Mini) motion picture still frame (grab), that is my base.
After the basic RAW footage has been trimmed to length there are several passages:
- Crop 21:9 aspect ratio (Cinemascope like)
- Levels: pretty much self explanatory and quite conservative, I increase the gamma and touch the in/out level marginally (more to be worked upon later)
- HF Color Temperature (it seems to introduce a bit of a reddish hue but I like it better than the Sony one)
- Color Corrector: this helps greatly not only balancing and enhancing the colours for highlights, midtones and shadows but also gaining more gamma without impacting the midtones details
- Color Curves: I use it to soften the highlights without flattening them or spacing them apart.
- Glow: used with discretion can help bring back the sky before grading it, use it too much and you're in for big artifacts
- HSL adjust: useful to give the last bit of shine to the picture
Then... I make a copy of it all, and use it as a base for the Sony Gaussian blur filter. That (source alpha, compositing child) track layer is applied selectively with an upper track layer (multiply mask, compositing parent) with a Sony Color gradient mask applied with Fx Mask generator (Alpha). This allows me to simulate the depth of field of a 35mm Cinema camera with Guassian filter of just 0.0015 to 0.0030 range (both horizontal and vertical).
The uppermost layer is just a regular alpha channel with a Color gradient mask to better grade wider regions (e.g. skies, dark expanses).
So far this worked ok for me and I am decently pleased with the result. I could surely improve, save for the fact that I am using outdated software, but I am far more interested in making videos rather than learning new techniques right now. I hope you can enjoy this insight into a stranger's workflow.
Last edited:
