Topaz has released the new Video AI 5.2 version, which includes the Rhea Enhancement Model, Frame Interpolation for After Effects, and Alpha layer support. The Rhea Enhancement Model combines Proteus and Iris to preserve fine details and handle text in a less destructive way. The model processes inputs at scale and then downscales to the selected output resolution. Video AI can now copy alpha layers from inputs and merge them back with AI model output videos. The new release also includes bug fixes and improvements such as restored preview set saving functionality, reduced UI lag, and correct adjustment of timeline width between different video inputs. Topaz will be launching Video AI Pro later this week, offering seat management, commercial usage rights, and multi-GPU support for teams. The previous implementation of multi-GPU rendering for the standard license will be discontinued [5e6f6bb5].
In addition to the new features and enhancements, Topaz has recognized that MFT cameras do not sell well in China and is planning to take action. OM Digital plans to continue the development of the PEN line, and a new OM Digital camera announcement is confirmed for 2024. Furthermore, a new Sigma patent reveals the design of two new MFT lenses: 20mm and 14mm f/1.4 [5e6f6bb5].