Mirrors/flowframes
1
0
Fork 0
mirror of https://github.com/n00mkrad/flowframes.git synced 2025-12-22 03:09:28 +01:00
Code Issues Packages Projects Releases Wiki Activity
489 Commits 1 Branch 0 Tags
b51ee375e5ce175a0921a9a77d90d0dc4deeb93b
Commit Graph

101 Commits

Author SHA1 Message Date
N00MKRAD
5556b5a5c7 Allow running CUDA on CPU (or unsupported NV GPUs) 2021-04-02 22:26:44 +02:00
N00MKRAD
b3ab84646b Switched to fractional framerates, fixed VFR inputs 2021-04-02 14:36:08 +02:00
N00MKRAD
e59fd53de7 Fixed wrong package version, minor logging improvements 2021-03-24 17:29:40 +01:00
N00MKRAD
7af4b980bb Fixed step-by-step frame renaming & added SBS scene blending support 2021-03-22 22:20:30 +01:00
N00MKRAD
abeb570f22 Fixed alpha interpolation with FLAVR 2021-03-14 15:26:52 +01:00
N00MKRAD
20d2922c29 Network checks no longer stop the window from rendering on startup 2021-03-13 14:48:19 +01:00
N00MKRAD
989fe15a09 Added FLAVR code, improved frame padding for FLAVR 2021-03-11 23:15:43 +01:00
N00MKRAD
692f6c9be9 Removed package manager remnants, added FLAVR network (WIP) 2021-03-11 12:58:18 +01:00
N00MKRAD
eebba121c9 Scene detection now works with frame sequence input, cleanup 2021-03-10 20:45:48 +01:00
Julian Maier
e70d72bc5a Fixed models.txt download deleting old file if it fails to get a new one 2021-03-05 10:38:29 +01:00
N00MKRAD
504242dca7 Revert to old UHD mode (RIFE CUDA) 2021-03-02 22:54:30 +01:00
N00MKRAD
d1d90d0419 try/catch blocks for AI functions 2021-03-01 18:13:19 +01:00
N00MKRAD
deb99070e4 RIFE CUDA scale arg from config 2021-02-28 16:34:47 +01:00
N00MKRAD
7563778a90 Don't delete models.txt before fetching (DownloadFile overrides anyway) 2021-02-28 11:57:02 +01:00
N00MKRAD
72cb1e93dc Update AI Model ComboBox after fetching models.txt 2021-02-28 11:56:09 +01:00
N00MKRAD
566546bef2 Merge pull request #27 from richardletshacks/github-model-mirror 
auto update model list and make github the default mirror
 2021-02-27 12:47:42 +01:00
N00MKRAD
f30ae16ce7 RIFE-CUDA image piping tests [WIP] 2021-02-27 12:46:25 +01:00
Richard
1d0e2a6c5e auto update model list 2021-02-26 23:25:27 +01:00
N00MKRAD
453787767c Check for CUDA availability before trying to run CUDA AI 2021-02-24 15:08:18 +01:00
N00MKRAD
8dd1d4ee4c NvApi UX improvements 2021-02-24 14:56:47 +01:00
N00MKRAD
0a71157a2e Capture and show peak fps out after interpolation 2021-02-23 14:05:58 +01:00
N00MKRAD
ff268f194f Changed Tensor Core check to include RTX Quadro cards 2021-02-23 12:21:54 +01:00
N00MKRAD
1d04d2d3ee RIFE-CUDA fp16 option and warplayer.py optimization 2021-02-21 11:26:18 +01:00
N00MKRAD
f491b6f6f9 Show error if trying to run FF out of an archive (temp folder) 
+ other UI improvements
 2021-02-18 14:30:56 +01:00
N00MKRAD
034794d5e5 RIFE-CUDA: Use more wthreads for higher factors, custom buffer size 2021-02-16 12:14:26 +01:00
Richard
70fe5291e9 fix progress 2021-02-11 21:05:45 +01:00
Richard
62b7650e69 better implementation that works for all three AIs 2021-02-11 12:33:09 +01:00
Richard
bc74f7e0ef implement faster progress updating for rife cuda 2021-02-11 09:05:10 +01:00
N00MKRAD
2e2664b8b3 Added RIFE-NCNN 2.0 workaround, fixed current.stepByStep wrong value 2021-02-10 23:16:04 +01:00
N00MKRAD
63da55cf6c Ensure AutoEnc is not paused from prev run when starting interpolation 2021-02-09 18:23:22 +01:00
N00MKRAD
babfe6e744 Revert "Convert to .NET 5 Project" 2021-02-09 16:07:57 +01:00
Dankrushen
41fedece97 Fix link buttons 2021-02-08 19:40:16 -05:00
N00MKRAD
e50c55ef90 Removed SemVer class and use Version class instead 2021-02-05 15:57:44 +01:00
N00MKRAD
d301991f1e Fixed SemVer comparison 2021-02-05 15:52:44 +01:00
N00MKRAD
2bdda6522f FfmpegCommands refactoring, renamed AudioVideo namespace 2021-02-02 12:56:48 +01:00
N00MKRAD
985ce45f1d Fixed ReverseRenaming being ultra slow + only run if SBS mode 2021-02-01 21:50:05 +01:00
N00MKRAD
2b1fa68418 serialize and save InterpSettings, float interpFactor variable 2021-02-01 18:05:50 +01:00
N00MKRAD
66bc4b557a Cleaner alpha extraction/merging 2021-01-30 13:14:57 +01:00
N00MKRAD
712f336acd Remove alpha interp frames after merging into rgb frames 2021-01-27 21:05:49 +01:00
N00MKRAD
2c872629fb Added alpha support to RIFE-NCNN 2021-01-27 14:29:49 +01:00
N00MKRAD
3710de3fb9 Added alpha support to DAIN-NCNN 2021-01-27 14:20:27 +01:00
N00MKRAD
80225b6f1c Cleaner RIFE NCNN recursion code 2021-01-27 12:04:22 +01:00
N00MKRAD
40e63fe51d Improved alpha support 2021-01-27 11:41:05 +01:00
N00MKRAD
86e27a4041 Only loop output if loop is on, fix SBS frame extraction not clearing, more 
Improved Interpolate.Cleanup() with retries
minOutVidLength now defaults to 5s instead of 0s
Frames filenameMap now uses zeropadding value from Padding class
 2021-01-26 16:45:02 +01:00
N00MKRAD
b5d3d9b07e Improved AI process error handling 2021-01-21 12:07:17 +01:00
N00MKRAD
55d13a3458 Better error handling for DAIN OOM 2021-01-21 11:50:31 +01:00
N00MKRAD
61aa0c40b8 Re-added DAIN, integrated into model downloader 2021-01-21 11:39:08 +01:00
N00MKRAD
8e868331d8 Added TTA (NCNN) config option, updated 3070 benchmarks 2021-01-20 20:24:46 +01:00
N00MKRAD
fdf62ff547 Fixed progress FPS being inf 2021-01-19 00:53:00 +01:00
N00MKRAD
ec5f494be6 Alpha support for GIFs/PNGs (no ncnn support atm) 2021-01-18 15:32:45 +01:00