Change the Camera Angle of Any Photo With AI: Qwen Image Edit Multiple Angles LoRA in SwarmUI
BiRefNet HR removes backgrounds at high resolution in single or batch mode, with BF16, FP16 or FP32 precision and a No Resize mode that keeps native size.
Published Nov 24, 2025 · 2,384 views · Watch on YouTube
Get the app and files

BiRefNet HR Background Remover: 1-Click Gradio App for Windows & Cloud
BiRefNet HR removes backgrounds at high resolution in single or batch mode, with BF16, FP16 or FP32 precision and a No Resize mode that keeps native size.
Chapters
- 0:00 Introduction to 3 Famous Qwen Image Edit LoRAs: Fusion, Relight, and Multiple Angles
- 0:41 Overview of Qwen Image Relight LoRA capabilities for changing natural lighting
- 0:59 Overview of Qwen Image Multiple Angles LoRA for changing camera perspectives
- 1:20 How to download and install the latest SwarmUI Model Downloader script
- 1:47 Verifying the new Qwen Image Edit LoRAs in SwarmUI Bundles list
- 2:19 Using the search feature to manually find and download specific Qwen LoRAs
- 2:48 How to update SwarmUI presets using the Windows Preset Delete & Import script
- 3:19 Setting up the Qwen Image Fusion LoRA workflow and selecting the correct preset
- 3:40 Extracting the required trigger word and prompt from LoRA metadata description
- 4:07 Introduction to BiRefNet HR High Resolution Background Remover tool
- 4:42 Reference to previous SwarmUI and ComfyUI backend installation tutorials
- 5:24 BiRefNet HR installation completion and GPU compatibility (RTX 5000/6000/Cloud)
- 5:45 How to use BiRefNet HR: Uploading images and understanding resolution settings
- 6:27 Comparing BiRefNet processing resolutions: Original Size vs 2048x2048
- 7:09 Handling VRAM errors in BiRefNet and viewing the high-quality background removal result
- 7:50 Preparing the composite image for Fusion using Paint.NET (Layers and Resizing)
- 8:35 Uploading the prepared image to SwarmUI and setting accurate aspect ratio
- 8:58 Explaining Smart Image Prompt Resizing and starting the Fusion generation
- 9:48 Reviewing the Qwen Image Fusion results and shadow generation quality
- 10:04 Comparing results with Smart Image Prompt Resizing enabled versus disabled
- 10:41 Setting up the Qwen Image Relight LoRA workflow and applying the preset
- 11:14 Uploading input image for Relight and configuring aspect ratio settings
- 11:50 Generating the Relight images and discussing VRAM usage on different GPUs
- 12:39 Setting up the Qwen Image Multiple Angles LoRA workflow and prompt selection
- 13:12 Uploading input image for Angle changing and adjusting resolution settings
- 13:33 Updates on Flux configurations, Torch Compile, and training speed improvements
- 14:56 Reviewing the Top-Down View results from the Multiple Angles LoRA generation
- 15:23 Conclusion and where to find support for Qwen Image Editing workflows
About this video
In this tutorial I am going to introduce you newest 3 Qwen Image Edit model LoRAs to fuse images (merge images naturally), relight images (change lighting of the scene) and also change camera angle (like from front view into top down view) of the images. With our 1-click to install and ready presets, you will get amazing quality images so easily without depending on OpenAI image model or Google's Gemini Nano Banana Pro model.
🔗 LINKS & RESOURCES MENTIONED
Step 1: SwarmUI Installer and Model Downloader
[➡️ https://www.patreon.com/posts/SwarmUI-Installer-Model-Downloader-114517862 ]
Step 2: BiRefNet HR (High Resolution) Background Remover
[➡️ https://www.patreon.com/posts/BiRefNet-HR-Installer-114517862 ]
Step 3: SwarmUI Windows Tutorial
[➡️ https://youtu.be/c3gEoAyL2IE ]
Step 4: SwarmUI Cloud Tutorial
[➡️ https://youtu.be/bBxgtVD3ek4 ]
Windows Requirements Tutorial (MANDATORY for first-time setup!)
[➡️https://youtu.be/DrhUHnYfwC0 ]
Video Chapters
0:00 Introduction to 3 Famous Qwen Image Edit LoRAs: Fusion, Relight, and Multiple Angles
0:41 Overview of Qwen Image Relight LoRA capabilities for changing natural lighting
0:59 Overview of Qwen Image Multiple Angles LoRA for changing camera perspectives
1:20 How to download and install the latest SwarmUI Model Downloader script
1:47 Verifying the new Qwen Image Edit LoRAs in SwarmUI Bundles list
2:19 Using the search feature to manually find and download specific Qwen LoRAs
2:48 How to update SwarmUI presets using the Windows Preset Delete & Import script
3:19 Setting up the Qwen Image Fusion LoRA workflow and selecting the correct preset
3:40 Extracting the required trigger word and prompt from LoRA metadata description
4:07 Introduction to BiRefNet HR High Resolution Background Remover tool
4:42 Reference to previous SwarmUI and ComfyUI backend installation tutorials
5:24 BiRefNet HR installation completion and GPU compatibility (RTX 5000/6000/Cloud)
5:45 How to use BiRefNet HR: Uploading images and understanding resolution settings
6:27 Comparing BiRefNet processing resolutions: Original Size vs 2048x2048
7:09 Handling VRAM errors in BiRefNet and viewing the high-quality background removal result
7:50 Preparing the composite image for Fusion using Paint.NET (Layers and Resizing)
8:35 Uploading the prepared image to SwarmUI and setting accurate aspect ratio
8:58 Explaining Smart Image Prompt Resizing and starting the Fusion generation
9:48 Reviewing the Qwen Image Fusion results and shadow generation quality
10:04 Comparing results with Smart Image Prompt Resizing enabled versus disabled
10:41 Setting up the Qwen Image Relight LoRA workflow and applying the preset
11:14 Uploading input image for Relight and configuring aspect ratio settings
11:50 Generating the Relight images and discussing VRAM usage on different GPUs
12:39 Setting up the Qwen Image Multiple Angles LoRA workflow and prompt selection
13:12 Uploading input image for Angle changing and adjusting resolution settings
13:33 Updates on Flux configurations, Torch Compile, and training speed improvements
14:56 Reviewing the Top-Down View results from the Multiple Angles LoRA generation