Stereo Conversion
Side-by-side 3D from any 2D source.
5
Output formats
~0.3s
Per frame warp + composite (1080p)
0 uploads
Desktop processing by default
Technical overview
Stereo conversion synthesizes left and right eye views from a depth map and the original frame. Each pixel is shifted horizontally based on its depth — closer objects move more, creating natural parallax that your brain interprets as 3D.
The hard part is not the shift — it is what happens at the edges. When a foreground object moves, it reveals background that was hidden in the original frame. anelo uses edge-aware inpainting to fill these disoccluded regions, producing clean results where naïve warping would leave gaps or stretching artifacts.
Disparity is comfort-checked by default. Excessive parallax causes eye strain and headaches. anelo enforces configurable comfort limits based on the target display type (VR headset vs. 3D TV vs. cinema screen), so output is watchable for extended sessions without fatigue.
Use cases
- Convert legacy 2D catalog titles for VR distribution
- Create immersive 3D versions of YouTube or social media content
- Build a stereo 3D highlight reel from 2D sports footage
- Generate 3D content for VR headset platforms without a stereoscopic camera
- Deliver MV-HEVC for Apple Vision Pro from standard 2D source
Configuration
- Baseline (inter-ocular distance)slider
Controls how wide the stereo separation is. Higher values produce a stronger 3D effect but risk comfort issues.
Default: 6.5 cm
- Convergenceslider
Sets the screen-plane depth — objects at this depth appear at the display surface. Objects closer pop out, objects farther recede.
Default: Auto
- Comfort limitselect
Maximum allowed disparity. Options tuned for different viewing contexts.
Default: VR headset (conservative)
- Output formatselect
Target stereo layout for the final file.
Default: Side-by-side (half)
Pipeline stages
Synthesizes left and right eye views from the depth map and source frame. Shifts pixels horizontally based on their depth to create a stereo pair with natural-looking parallax.
Configurable baseline (inter-ocular distance) and convergence. Edge inpainting fills disoccluded regions.
Combines left and right eye views into the selected stereo output format. Supports side-by-side, top-bottom, anaglyph, and per-eye delivery for VR headsets.
Optional vertical squeeze for half-SBS. Per-eye output for maximum quality.
Available models
Output formats
Full resolution left + right. Best quality. Double the file size of the original.
Each eye at half horizontal resolution. Compatible with most 3D TVs and VR players.
Stacked stereo layout. Common in broadcast and some VR applications.
Viewable with red/cyan glasses. Useful for quick previews — not recommended for production.
Apple Multiview HEVC for Vision Pro. Native spatial video playback.
Workflows
Full 2D-to-3D pipeline
Source video runs through depth estimation, stereo warping, and compositing in one pass. The output is a ready-to-watch SBS, top-bottom, or MV-HEVC file. No intermediate steps required.
Batch season conversion
Queue an entire season of episodes for overnight processing. Preflight scoring filters problematic frames. Each episode outputs as a separate file with consistent stereo parameters across the batch.
Format export for multiple platforms
Run the same source through the pipeline once, then export to multiple formats: half-SBS for streaming, full-SBS for archival, and MV-HEVC for Apple Vision Pro — all from the same depth data.
Related products
View allIndustries using this
Start with a free desktop install.
Desktop processing is free with no job limits. Pro adds cloud processing, watermark removal, and advanced 3D controls.