Oct 24, 2023
V7 can now support frame-accurate video files that are 10x longer and accommodate 700x more annotations than before. While users would previously split a 90 minute football game into multiple shorter clips, such a video can now be annotated performantly as one file within V7.
On top of that, we optimized back-end file processing to reduce potential storage costs. If your datasets use custom storage options, you can now expect up to a 90% reduction in imported and processed file size with no reduction in quality.
10x longer video files (up to c.100K frames)
Up to 700x more simultaneous video annotations (assuming they are equally distributed across the video)
90% less storage consumed for imported footage
Since you can now import files of up to and beyond 1 hour in length at their native frame rate, you no longer have to slice long videos into multiple shorter clips and then stitch them back together again. This is also a big boost for annotator efficiency as they now maintain context over the entire video at once.
Andrea Azzini
Head of Product
