User Guide

System Requirements

Minimum Requirements

• macOS: 14.0 (Sonoma) or later
• Chip: Apple Silicon (M1, M2, M3, M4 or later)
• RAM: 8GB (16GB recommended for 4K+ content)
• Storage: 2GB for app + models; additional space for exports

For MV‑HEVC Spatial Video Export

• macOS: 15.0 (Sequoia) or later
• Chip: M2 or later with hardware MV‑HEVC encoder support

Supported Input Formats

Any video format macOS can decode via AVFoundation:

• MP4, MOV, M4V, MKV, AVI, WebM
• H.264, H.265/HEVC, ProRes, VP9
• Up to 8K resolution
• Variable and constant frame rates

Interface Overview

Main Window Layout

• Files Sidebar (Left) — Your import queue. Drag videos here or click + to add files. Click a file to select it for preview/export.
• Preview Area (Center) — Displays the current frame with selected view mode. Scrub the timeline below to navigate.
• Controls Panel (Right) — Output mode, depth model, quality settings, and export controls.
• Timeline (Bottom) — Playback controls, frame scrubber, and In/Out point markers.

Preview Modes

• Video — Original source footage
• Depth Map — Grayscale depth visualization (white = near, black = far)
• Anaglyph — Red/cyan stereo preview for quick 3D verification
• Side‑by‑Side — Left/right eye preview (Pro)

Status Bar

The bottom status bar shows current frame number, timecode, resolution, and processing status during export.

Quick Start

1. Add videos by dragging files into the app or clicking + in the Files sidebar.
2. Select an output mode (Vision Pro / SBS / Top‑Bottom / Anaglyph / Separate L/R).
3. Choose a depth model and adjust Depth / Bias / Smooth.
4. Preview in Video / Depth Map / Anaglyph.
5. (Optional) Set an In/Out range to export only part of the clip.
6. Click Convert (selected clip) or Convert All (batch).

Output Modes

• Vision Pro (MV‑HEVC) — Spatial video .mov for Apple Vision Pro (when supported by your Mac).
• Side‑by‑Side (SBS) — Stereo format for Quest, YouTube VR, and VR headsets. Auto‑switches to H.264/MP4.
• Top‑Bottom (TB) — Stereo format (stacked vertically) for VR players. Auto‑switches to H.264/MP4.
• Anaglyph — Red/cyan stereo for quick depth checks or legacy displays.
• Separate Left/Right — Exports two files (LEFT + RIGHT) for advanced workflows.
• Image sequence — Export frames as PNG/TIFF/EXR (useful for VFX pipelines).
• RGBD / depth‑in‑alpha — Packs depth into the alpha channel (advanced pipelines).

Choosing the Right Format

Depth Models

Immersive Converter uses state‑of‑the‑art AI models to estimate depth from 2D video frames.

Available Models

• Depth Pro (CoreML) — Apple's flagship monocular depth model. Produces sharp, metric‑accurate depth maps with excellent edge detection. Runs natively on Apple Neural Engine. Included in Standard Edition.
• DA V2 Large (CoreML) — Alternative model optimized for different scene types. May perform better on certain content like landscapes or architecture. Included in Standard Edition.
• Depth Pro (ONNX) — Same model running via ONNX Runtime with GPU acceleration. Useful for comparison or specific hardware configurations. Pro Edition only.
• Depth Anything V2 (Python) — Full Python implementation with additional configuration options. Pro Edition only.

Model Selection Guidelines

• General footage: Start with Depth Pro (CoreML) — fastest and best overall quality.
• Outdoor/landscape: Try DA V2 Large if Depth Pro struggles with distant objects.
• High detail needed: Depth Pro typically produces sharper edges.
• Noisy/low-light footage: Increase Smooth slider regardless of model choice.

Preview & Playback

• Video — Normal playback of the source.
• Depth Map — View the estimated depth.
• Anaglyph — A fast "stereo sanity check" while tuning depth.

Changing any setting automatically re‑renders the preview so you can see results instantly.

Trim Range (In/Out Points)

You can export only part of a video by setting an In/Out range:

Depth Controls

Fine‑tune the 3D effect to match your content and viewing preferences.

Basic Controls

• Depth (0–100%) — Overall stereo separation intensity. Higher values create stronger 3D effect but may cause discomfort. Start around 50% and adjust based on content.
• Bias (‑100 to +100) — Shifts the zero‑parallax plane (the depth where objects appear "at screen level"). Negative values push content into the screen; positive values pull content toward the viewer.
• Smooth (0–100%) — Temporal smoothing to reduce depth flicker between frames. Essential for noisy or low‑light footage. Higher values may slightly soften depth transitions.

Pro Advanced Controls

• Parallax Clamp — Limits maximum stereo separation to prevent eye strain. Enable for comfortable extended viewing.
• Edge Feathering — Softens depth transitions at object boundaries to reduce "cardboard cutout" artifacts.
• Temporal Denoise — Advanced frame‑to‑frame depth consistency. More effective than basic Smooth for complex motion.
• Depth Range (Near/Far) — Manually set the depth range if auto‑detection produces unexpected results.

Quick Presets

• Make Comfortable — Auto‑enables safe depth clamping and moderate settings for extended viewing.
• Cinematic — Subtle, film‑like depth suitable for narrative content.
• Immersive — Stronger depth for maximum 3D impact (best for short clips).

Quality, Resolution & Frame Rate

Quality Presets

• Draft — Fastest processing, lower depth accuracy. Good for quick previews and testing settings.
• Standard — Balanced quality and speed. Suitable for most content.
• High — Better depth accuracy, slower processing. Recommended for final exports.
• Pro — Maximum quality, highest accuracy. Best for professional work and detailed scenes.

Resolution Options

• Source — Maintains original resolution (up to 8K)
• 1080p — Full HD, fastest export
• 1440p — QHD, good balance for VR
• 2160p (4K) — Ultra HD, recommended for Vision Pro
• Custom — Specify exact dimensions (Pro)

Frame Rate

• Auto (Source) — Preserves original timing. Best for variable frame rate (VFR) content like iPhone videos.
• 24/25/30/60/90 fps — Converts to constant frame rate (CFR).
• Frame Blending — Enable when increasing frame rate for smoother motion interpolation.

Upscaling (Pro)

Pro Edition includes AI upscaling up to 2× resolution:

• 1.5× — Moderate enhancement with minimal artifacts
• 2.0× — Maximum upscale (e.g., 1080p → 4K)

Batch Processing

Process multiple videos with the same settings automatically.

Adding Files to Queue

• Drag multiple files into the Files sidebar
• Use ⌘ + O and select multiple files
• Drag an entire folder to add all videos inside

Batch Export

1. Add all videos to the queue
2. Configure your desired output mode, depth settings, and quality
3. Click Convert All (instead of Convert)
4. All videos process sequentially with identical settings

Queue Management

• Reorder: Drag files in the sidebar to change processing order
• Remove: Right‑click → Remove, or select and press Delete
• Clear All: Right‑click → Clear Queue

Where Exports Are Saved

• Click Open Exports to reveal the current export folder in Finder.
• Default location: ~/Movies/Immersive Converter/
• You can set a custom export folder in Settings → Export Location.

File Naming

Exported files follow the pattern: OriginalName_[FORMAT]_[QUALITY].ext

Example: vacation_SBS_High.mp4

Direct to Photos (Vision Pro)

For MV‑HEVC spatial videos, you can optionally export directly to Photos.app for automatic iCloud sync to Vision Pro.

Keyboard Shortcuts

Playback

Editing

View

File Operations

Pro Features

Pro Edition unlocks advanced capabilities for professional workflows.

Additional Depth Models

• DA V2 Large — Higher capacity model for complex scenes
• ONNX Runtime — Alternative inference engine with GPU acceleration
• Custom Models — Load your own ONNX/CoreML depth models

Advanced Depth Controls

• Parallax Clamping — Limit maximum stereo separation for viewer comfort
• Edge Feathering — Smooth depth transitions at object boundaries
• Temporal Consistency — Advanced frame‑to‑frame depth smoothing
• Manual Depth Range — Override auto‑detected near/far planes
• Depth Curve — Custom gamma/curve adjustment for depth distribution

Import External Depth Maps

Bypass AI estimation by loading pre‑computed depth:

• Grayscale video file (same resolution/frame count as source)
• Image sequence (PNG, TIFF, EXR)
• Useful for VFX pipelines with depth from Nuke, Blender, or other tools

Python Model Setup

Pro builds can use additional Python‑based AI models:

1. Open AI Model Setup from the right panel
2. Click Install Environment to set up isolated Python environment
3. Use Download Models to fetch model weights (may be several GB)
4. Select Python models from the model dropdown once installed

Priority Support

Pro license includes priority email support with faster response times and direct access to the development team for complex issues.

Audio

Immersive Converter preserves audio from your source files during export.

Supported Audio

• AAC, MP3, PCM/WAV audio tracks are preserved
• Multi‑channel audio (5.1, 7.1) passes through unchanged
• Spatial audio metadata is preserved when present

Audio Sync

Audio remains synchronized with video during:

• Trim/In‑Out exports
• Frame rate changes (with slight adjustment for timing)
• Resolution changes

Troubleshooting Audio

• If audio is missing, try MOV container instead of MP4
• Some codecs may not support audio passthrough — re‑encode with AAC
• Verify source file actually contains audio (some screen recordings don't)

Tips & Best Practices

Getting the Best Depth Quality

• Good lighting matters: Well‑lit footage produces more accurate depth maps. Low‑light or high‑ISO footage may need more Smooth.
• Avoid flat scenes: The AI needs visual cues. Plain walls or solid colors produce poor depth. Textured surfaces work best.
• Motion helps: Camera movement and parallax give the AI more information. Static locked‑off shots may have less accurate depth.
• Faces are special: Depth Pro is trained to handle faces well. Portrait content typically converts beautifully.

Comfortable 3D Viewing

• Less is more: Subtle depth often looks more natural than aggressive 3D. Start with Depth at 30–50%.
• Mind the edges: Objects at frame edges with extreme depth can cause discomfort. Use Edge Feathering (Pro).
• Convergence matters: Adjust Bias so the main subject sits near the screen plane (zero parallax).
• Test on device: What looks good in anaglyph preview may feel different on Vision Pro. Always test final exports.

Optimizing Performance

• Use Draft for testing: Dial in your settings with Draft quality, then switch to High/Pro for final export.
• Close other apps: AI inference benefits from available GPU/Neural Engine resources.
• Proxy workflow: For very long projects, convert a lower‑res proxy first, finalize settings, then export full resolution.
• SSD storage: Export to fast SSD storage. Slow drives bottleneck video encoding.

Content That Works Best

• ✓ Travel videos with depth and scenery
• ✓ Family videos and events
• ✓ Nature and wildlife footage
• ✓ Architecture and interiors
• ✓ Sports and action with clear subjects
• ✗ Screen recordings and flat UI content
• ✗ Heavily compressed or artifacted video
• ✗ Abstract or psychedelic visuals

Vision Pro Specific Tips

• Export at native resolution (4K recommended) for best clarity
• Use "Make Comfortable" preset for extended viewing sessions
• Export to Photos for automatic iCloud sync
• Test short clips before committing to long exports

Troubleshooting

Export Fails Immediately (Vision Pro / MV‑HEVC)

• Your Mac may not support MV‑HEVC stereo encoding.
• Requires macOS 15.0+ and M2 chip or later.
• Check System Information → Graphics/Displays for hardware encoder support.
• Try exporting SBS or Top‑Bottom instead.

Export Is Empty / Shorter Than Expected

• Clear In/Out points with X and retry.
• Check that the In point is before the Out point.
• Verify the source file isn't corrupted (try playing in QuickTime).

3D Feels Too Strong / Uncomfortable

• Reduce Depth slider to 30–50%.
• Adjust Bias to move the convergence plane closer to main subject.
• (Pro) Enable Parallax Clamp or use Make Comfortable preset.
• Some content (like extreme close‑ups) may never feel comfortable in 3D.

Flicker / Unstable Depth

• Increase Smooth slider to 40–60%.
• Try a higher Quality preset.
• (Pro) Enable Temporal Denoise for advanced stabilization.
• Noisy/low‑light footage naturally produces less stable depth.

No Audio in Export

• Confirm the source file contains audio (check in QuickTime or VLC).
• Try MOV container instead of MP4.
• Some image sequence exports don't include audio by design.

"Cardboard Cutout" Effect

• This happens when depth has sharp edges between foreground/background.
• (Pro) Enable Edge Feathering to soften transitions.
• Reduce overall Depth intensity.
• Some content types (like green screen footage) may exhibit this regardless.

Depth Map Looks Wrong

• Try switching between Depth Pro and DA V2 Large models.
• Reflective surfaces (mirrors, glass, water) confuse depth estimation.
• Transparent objects may not depth correctly.
• Very uniform surfaces (white walls, sky) have ambiguous depth.

App Crashes During Export

• Reduce resolution or quality preset.
• Close other memory‑intensive applications.
• Check available disk space (exports need significant temporary space).
• Try exporting a shorter clip to isolate the problematic section.
• Report persistent crashes to support with macOS version and crash log.

Preview Is Slow / Laggy

• Use Draft quality for real‑time preview.
• Reduce preview resolution in Settings.
• 4K+ content is demanding — consider proxy workflow.
• Ensure no thermal throttling (Mac getting hot).

Import Fails / File Not Recognized

• Verify the file plays in QuickTime Player.
• Try transcoding to ProRes or H.264 first.
• Some obscure codecs may not be supported.
• Corrupted files will fail to import.

Glossary

Anaglyph

A stereoscopic technique using color filters (typically red/cyan) to display 3D images on regular 2D screens. Requires colored glasses to view.

Bias / Convergence

The depth plane where left and right eye images align perfectly (zero parallax). Objects at this plane appear "at screen level."

CoreML

Apple's machine learning framework that runs AI models efficiently on Apple Silicon using the Neural Engine, GPU, and CPU.

Depth Map

A grayscale image where brightness represents distance from camera. White = close, black = far. Used to generate stereoscopic 3D.

Depth Pro

Apple's state‑of‑the‑art monocular depth estimation model that produces metric‑accurate depth from single images.

MV‑HEVC

Multi‑View High Efficiency Video Coding. Apple's format for spatial video that encodes left and right eye views efficiently.

Monocular Depth

Estimating 3D depth from a single 2D image (one "eye"), as opposed to stereo depth which uses two cameras.

Parallax

The apparent displacement of objects when viewed from different positions. In stereo 3D, the horizontal offset between left and right eye views.

ProRes 4444

Apple's professional video codec that supports alpha channel (transparency). Required for RGBD depth‑in‑alpha exports.

RGBD

RGB + Depth. A format that combines color video with depth information, typically stored in the alpha channel.

SBS (Side‑by‑Side)

A stereo 3D format where left and right eye views are placed horizontally adjacent in a single frame.

Spatial Video

Apple's term for stereoscopic 3D video designed for Apple Vision Pro, typically encoded in MV‑HEVC format.

Stereoscopic / Stereo 3D

Creating the illusion of depth by presenting slightly different images to each eye, mimicking human binocular vision.

TB (Top‑Bottom)

A stereo 3D format where left and right eye views are stacked vertically in a single frame.

Temporal Smoothing

Processing that ensures consistency between consecutive video frames, reducing flicker and jitter in depth maps.

Zero Parallax Plane

The depth at which objects appear to be at screen level (neither popping out nor receding). Also called the convergence plane.

Privacy

• Video processing and AI depth estimation run locally on your Mac.
• Standard builds are intended to work offline after install.
• Pro may download AI model files if you choose to enable Python/GPU models.

For full details, see our Privacy Policy.

Support

Getting Help

We're here to help you get the most out of Immersive Converter.

• Email: [email protected]
• Response time: Standard within 48 hours, Pro within 24 hours

Before Contacting Support

• Check this guide and the FAQ for common solutions
• Try restarting the app and your Mac
• Ensure you're running the latest version
• Test with a different video file to isolate the issue

When Reporting a Problem

To help us resolve your issue quickly, please include:

• System info: macOS version (e.g., 15.1) + Mac model (e.g., MacBook Pro M2)
• App version: Found in Immersive Converter → About
• Input file: Format (MOV/MP4), resolution, frame rate, duration
• Export settings: Output mode, quality preset, depth settings
• Steps to reproduce: What exactly triggers the problem?
• Error messages: Screenshot or exact text of any error dialogs
• Crash log: If the app crashes, include the crash report from Console.app

Feature Requests

We love hearing your ideas! Send feature requests to [email protected]

Refunds

Purchases are handled through the Mac App Store. For refund requests, contact Apple directly through reportaproblem.apple.com within 14 days of purchase.