Controls / Flip Depth
Reverses the interpretation of near and far deep samples.
← Fusion node reference
Deep CompositingIntermediate
DeepToImage
Deep image to standard 2D image
Converts a Deep Image back into a conventional 2D Fusion image after depth-aware compositing is complete.
DeepToImage
DaVinci Resolve · Fusion
DeepToImage is normally the final node in a Deep compositing branch. It resolves the depth samples and optional volumetric information into an ordinary image stream that can continue through ColorCorrector, Glow, Blur, Merge or MediaOut.
Overview
What DeepToImage does
DeepToImage is normally the final node in a Deep compositing branch. It resolves the depth samples and optional volumetric information into an ordinary image stream that can continue through ColorCorrector, Glow, Blur, Merge or MediaOut.
Mental model
Think of DeepToImage as flattening a depth-aware stack only after all operations that need multiple samples per pixel have finished.
Common uses
Finish a Deep compositing chain
Convert Deep renders before 2D grading
Resolve a DeepMerge result
Flatten volumetric Deep passes
Return to ordinary Fusion image processing
Flow anatomy
Inputs & outputs
The port structure shows how this node fits into a Fusion graph.
Input
A Deep Image containing depth samples and optional volumetric information.
DeepToImage
Fusion node
Output
A standard 2D Fusion image with the Deep structure resolved.
Inspector
Important parameters
Start with these controls before reaching for more complicated workarounds.
Controls / Volumetric Composition
Enables volumetric compositing when deep samples contain volume information.
Image / Process Mode
Selects the processing mode.
Chooses how the node evaluates or combines the source data. Available modes include Full Frames.
Image / Depth
Chooses the image depth/bit-depth used for the flattened DeepToImage output.
Source Color Space / Color Space Type
Sets the color Space Type.
Source Gamma Space / Curve Type
Selects the Curve Type used by DeepToImage.
Source Gamma Space / Remove Curve
Removes the selected gamma or transfer curve so the source is treated in a linearized form.
Source Gamma Space / Pre-Divide / Post-Multiply
Controls premultiplied-alpha handling around the operation.
Pre-divide removes alpha multiplication before processing; post-multiply restores it afterward. This is useful when color work creates dark or bright fringes around translucent edges.
Practical setup
A reliable workflow
- 1
Complete all DeepMerge / DeepHoldout work first.
- 2
Check whether depth direction is correct.
- 3
Enable volumetric composition for smoke, fog or transparent volumes.
- 4
Choose an appropriate floating-point output depth.
- 5
Continue with normal 2D nodes after DeepToImage.
Typical node chain
Deep Image→
Deep compositing→
DeepToImage→
2D image
Examples
Where this node fits in real work
Finish a Deep composite
Flatten the depth-aware result before final grading.
dMergedHoldoutDeepToImageColorCorrectorMediaOut
Volumetric pass
Resolve a Deep fog or smoke pass and then add final glow.
Deep ImageDeepToImageGlowMediaOut
Troubleshooting
Common mistakes
Flattening the Deep stream too early
Ignoring reversed Z conventions
Using low bit depth before heavy grading
Disabling volumetric composition on true volume data
Production notes
Practical tips
Keep DeepToImage as late as possible in the Deep branch.
Use float16/float32 for VFX work.
If occlusion looks inverted, test Flip Depth before rebuilding the comp.
Related nodes