Another video mode, this time will accept dual mp4 videos shot in fisheye.
GoPro Fusion cameras also produce a dual .mp4 video format (two fisheye videos that can be stitched together to create a single equirectangular 360).
For GoPro Fusion Unprocessed Video mode (dual mp4s):
takes GoPro two GOPro fisheye videos with GPMF data
converts to an equirectangular frames
optionally adds a nadir to the resulting video
stores output/uploads output/etc
GoPro2GSV validates the 2x .mp4 inputs before processing as follows;
file extension must be .mp4
filenames must start with GPFR and GPBK (there must be one of each)
GoPro:DeviceName is Back Lens
GoPro:DeviceName is Front Lens
Front video (filename starts with GPFR) trackN:GPSDateTime has more than 10 entries.
Track1:SourceImageWidth is the same in both videos
Track1:SourceImageHeight is the same in both videos
For processing, the same flags as .mp4 video mode can be used, and the processing is somewhat similar to mp4.
ffmpeg first extracts frames from each video. However, corresponding frames from each video need to be joined converted to equirectangular projection.
This will be the same as #31 from this point
In this mode, a user should be able to set --keep_extracted_frames which will also store a copy of the extracted images used to create final video (with geotags). If not set, frames extracted will be removed on completion.
===== below docs
dual fisheye mp4 video -> timelapse frames -> final video (for Fusion cameras)
this mode takes two fisheye mp4 videos, breaks each video into frames at a user defined frame rate, stitches them together into equirectangular frames using fusion2sphere, removes erroneous GPS points/frames, and the extracted frames are then rebuilt into an equirectangular video, or set of equirectangular videos. The user options are the same as equirectangular mp4 video -> timelapse frames -> final video mode
Another video mode, this time will accept dual mp4 videos shot in fisheye.
GoPro Fusion cameras also produce a dual .mp4 video format (two fisheye videos that can be stitched together to create a single equirectangular 360).
For GoPro Fusion Unprocessed Video mode (dual mp4s):
GoPro2GSV validates the 2x
.mp4
inputs before processing as follows;.mp4
GPFR
andGPBK
(there must be one of each)GoPro:DeviceName
isBack Lens
GoPro:DeviceName
isFront Lens
GPFR
)trackN:GPSDateTime
has more than 10 entries.Track1:SourceImageWidth
is the same in both videosTrack1:SourceImageHeight
is the same in both videosFor processing, the same flags as .mp4 video mode can be used, and the processing is somewhat similar to mp4.
ffmpeg first extracts frames from each video. However, corresponding frames from each video need to be joined converted to equirectangular projection.
This will be the same as #31 from this point
In this mode, a user should be able to set --keep_extracted_frames which will also store a copy of the extracted images used to create final video (with geotags). If not set, frames extracted will be removed on completion.
===== below docs
dual fisheye mp4 video -> timelapse frames -> final video (for Fusion cameras)