Frame Interpolation
What is Frame Interpolation?
Frame interpolation creates new frames between existing ones to make video look smoother, like converting 24 FPS footage to 60 FPS or creating slow-motion from normal-speed video.
At a glance
- Also known as
- Motion interpolationFrame synthesisTemporal upscaling
- Used for
- Increasing video frame rate for smoother playbackCreating slow-motion from standard frame rate footageSmoothing AI-generated video output
- Common tools
- RIFEDAINTopaz video AIDaVinci resolve optical flowAfter effects frame blending
- Related terms
- Frame rateFrames per second (FPS)Optical flowSlow motionAI video generation
Ready to create?
Direct scenes, design characters, and ship full films
All-in-one AI creative platform with simple, transparent pricing, no speed throttles, and an infinite Canvas for max creativity.
How it compares
High frame rate capture records more frames per second at the point of filming, preserving the actual visual content of each moment. Frame interpolation synthesises new frames that were never actually recorded, creating plausible but ultimately artificial in-between states. Captured high frame rate footage is always higher quality than interpolated footage for equivalent frame rates, but requires appropriate hardware and adds storage and processing demands at the capture stage. Interpolation is the more accessible option when high frame rate capture was not used or was not possible.
Think of it like…
Think about a simple flipbook animation where you have drawn a figure at position one on page one and the same figure at position three on page three, but page two is blank. Without the middle page, the flip makes the figure jump suddenly from position one to three. Frame interpolation is like having a clever artist look at pages one and three and automatically draw what page two should look like to make the motion smooth. The artist figures out where everything was moving and fills in the missing moment. Sometimes the artist gets it right; sometimes, if the movement is very complicated, the drawing looks a bit strange: that is when interpolation artefacts appear.
Pro tip
Frame interpolation works best on footage with smooth, consistent motion and fails most visibly on very fast motion, complex background-foreground interaction, and scene cuts. Applying interpolation across edit cuts: where the model attempts to synthesise frames between two completely different images: produces severely artefacted results. In post-production workflows, applying interpolation to individual clips before assembly rather than to the assembled timeline prevents this common and visually disruptive error.
Types and variations
- Optical flow interpolation is the most sophisticated approach, using motion vector analysis to synthesise plausible in-between frames.
- Frame blending is a simpler method that averages adjacent frames to create intermediate content without motion analysis, producing a characteristic blurred look.
- AI-based interpolation using dedicated deep learning models like RIFE produces higher quality results than traditional optical flow methods, particularly for complex or fast motion.
- Depth-aware interpolation models like DAIN incorporate depth estimation to handle occlusion more accurately, improving results in scenes with foreground-background interaction.
- Real-time interpolation is implemented in display hardware and graphics drivers to smooth video playback on monitors and televisions.
Ready to make your first scene in Morphic?
Try MorphicCommon use cases
- Post-production pipelines use frame interpolation to create slow-motion sequences from standard-speed footage without requiring high-frame-rate capture.
- Video restoration workflows apply interpolation to archive footage to produce smoother, more modern-feeling playback.
- AI video generation workflows use interpolation as a post-processing step to smooth the motion in generated clips that were produced at lower frame rates.
- Content creators working for platforms that benefit from higher frame rate playback use interpolation to convert 24 FPS edited content to 60 FPS for upload.
- Game cinematics and animation studios use interpolation to create fluid motion between keyframes without hand-animating every intermediate position.
Ready to create?
Direct scenes, design characters, and ship full films
All-in-one AI creative platform with simple, transparent pricing, no speed throttles, and an infinite Canvas for max creativity.
FAQs
Frame interpolation is a technique that generates new synthetic frames between existing frames in a video sequence, increasing the frame rate and creating smoother motion. Modern AI-based methods analyse motion vectors and visual content to synthesise plausible in-between frames, producing results significantly superior to older techniques that simply blended adjacent frames.
AI frame interpolation uses neural networks trained on large amounts of video footage to estimate the motion occurring between adjacent frames and synthesise new image content that represents a plausible intermediate position along that motion. The process involves optical flow estimation: determining the direction and speed of every region's movement: followed by frame synthesis that generates the new image content at the interpolated temporal position.
Interpolation artefacts are visual errors that appear in synthesised frames when the model's motion estimation is incorrect or insufficient for the complexity of the content. Common artefacts include ghosting, where a translucent second image of a moving subject appears alongside the primary subject; smearing, where fast-moving objects leave a blur trail along their motion path; and incoherent blending, where regions of incompatible image content from adjacent frames are improperly merged.
RIFE, which stands for Real-Time Intermediate Flow Estimation, is one of the most widely used AI frame interpolation models. It is valued for its combination of high interpolation quality and fast processing speed, making it practical for post-production workflows rather than requiring prohibitive render times. RIFE is available as a standalone tool and is integrated into several professional video upscaling and enhancement applications including Topaz Video AI.
Frame interpolation can create convincing slow-motion effects from standard frame rate footage by generating additional synthetic frames between each real capture, stretching the footage temporally without the blur associated with simply slowing down footage below its native frame rate. For moderate slow-motion factors, AI interpolation produces high-quality results. Very high slow-motion factors require many synthetic frames per real frame, which increases the visibility of artefacts, particularly in complex or fast-moving scenes.
Frame interpolation is beneficial for AI-generated video that has visible temporal jitter or insufficient smoothness at its native frame rate. For footage that is already smooth, or for content with intentional cinematic 24 FPS aesthetics, interpolation may not improve and could actually change the visual character in undesirable ways by removing the natural motion blur associated with lower frame rates. Evaluating whether the specific footage benefits from smoothing before applying interpolation is the recommended approach.
Frame interpolation produces the best results on footage with smooth, consistent, relatively slow motion: landscapes, gentle camera movements, slow character movement. It performs less well on very fast motion, complex foreground-background occlusion, rapid camera pans, and scene transitions. Understanding these strengths and limitations helps creators apply interpolation selectively where it improves footage rather than applying it universally and introducing artefacts into scenes that did not need smoothing.
The soap opera effect is what many viewers call the unnaturally smooth, hyper-real appearance of films or television content that has been processed with frame interpolation: typically on consumer televisions with motion smoothing features enabled by default. The effect is named for its association with the higher frame rate aesthetics of daytime television drama. Whether the soap opera effect is desirable depends entirely on the content and context; it is considered inappropriate for theatrical film content by most filmmakers but preferred by some viewers for sports and other high-motion programming.
