Releases: NVIDIA/MAXINE-VFX-SDK
Releases · NVIDIA/MAXINE-VFX-SDK
v0.7.2.0. Release
v0.7.1.0 Release
Changelog (v0.7.1.0)
- NvCVImage_Transfer() now sets alpha to 255 or 1.0f when doing RGB -> RGBA
- NvCVImage_CompositeRect() has a premultiplied alpha mode added
- AI Green Screen filter
- Introduced NVVFX_MAX_INPUT_WIDTH, NVVFX_MAX_INPUT_HEIGHT parameter selectors and getters to set the maximum input resolution
for NVVFX_FX_GREEN_SCREEN pipeline to avoid lazy memory allocations - Introduced the following parameter selectors and APIs for NVVFX_FX_GREEN_SCREEN to improve the quality of the model
- NVVFX_MAX_NUMBER_STREAMS parameter selector to set the maximum number of streams
- NVVFX_STATE, NVVFX_STATE_COUNT parameter selectors
- NvVFX_AllocateState, NvVFX_DeallocateState, NvVFX_ResetState, NvVFX_SetStateObjectHandleArray APIs
- BatchEffectApp does not support NVVFX_FX_GREEN_SCREEN anymore. Use BatchAigsEffectApp as the sample app for batch input
- Introduced NVVFX_MAX_INPUT_WIDTH, NVVFX_MAX_INPUT_HEIGHT parameter selectors and getters to set the maximum input resolution
- Migrated to TensorRT 8.4.2.2
- Migrated to CUDA 11.6u1
v0.6.5.2 Release
Changelog (v0.6.5.2)
- AI Green Screen filter
- Quality mode is updated for higher quality
- Added CUDA graph optimization which may improve overall latency under a GPU-intensive workload.
- Super Resolution filter
- 4K input support for 4/3x(~1.33x), 1.5x and 2x super resolution
- Migrated to TensorRT 8.0.1.6
- Migrated to CUDA 11.3u1
v0.6 Beta
v0.5
v0.5 (Beta):
- Video Super Res network beta release:
-- Applies an AI Upscale and cleans the content of encoding artifacts
-- Supports 1.33x, 1.5x and 2x AI scaling
-- Supported input resolutions: 720p and 1080p - 2 performance modes:
-- High quality mode: SuperRes
-- High Performance mode: Upscale - Support for GeForce RTX 30 and RTX 20 Series, TITAN RTX, and Quadro RTX GPUs