Duplicate from akhaliq/Qwen-Image-2.1-Multiple-Angles-LoRA

Co-authored-by: AK <akhaliq@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,47 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/example_1.jpg filter=lfs diff=lfs merge=lfs -text
+examples/example_3.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_0.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_1.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_2.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_3.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_4.jpg filter=lfs diff=lfs merge=lfs -text
+examples/real_5.jpg filter=lfs diff=lfs merge=lfs -text
+examples/cat_input.jpg filter=lfs diff=lfs merge=lfs -text
+examples/dog_compare.gif filter=lfs diff=lfs merge=lfs -text
+examples/cat_compare.gif filter=lfs diff=lfs merge=lfs -text
+examples/orbit_cat.gif filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,176 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen-Image-2.1
+tags:
+  - qwen-image-2.1
+  - lora
+  - camera-control
+  - multiple-angles
+  - image-editing
+  - multi-view
+  - dome-objaverse
+library_name: diffusers
+pipeline_tag: image-to-image
+---
+# Qwen-Image-2.1 · Multiple-Angles camera-control LoRA
+
+Reshoot any object from a controlled camera position: feed a photo, ask for a different
+azimuth/elevation, get the same subject from that angle. First camera-control LoRA for
+**Qwen-Image-2.1** — the newest Qwen image base (unified generate + edit, native RGBA), where this
+format did not exist yet.
+
+- **Trigger**: `<mva>` — every caption/prompt starts with it
+- **36 poses**: 12 azimuths (30° steps) × 4 elevations (0° eye-level / 30° elevated / 60° high-angle / 90° top-down), plus a `close-up` variant of each → 72 addressable framings
+- **Recommended checkpoint**: step 1,000 (`checkpoints/…_000001000.safetensors`) — angle-following, identity preservation and detail peak together there; step 1,500 is the alternate, step 2,500 bakes harder (drifts toward copying the reference view)
+- **Strength**: 0.8–1.0 in ComfyUI / diffusers
+
+## Real photos
+
+In-distribution examples are above — but does it work on everyday photos it never saw? Dog, cat and
+apple from Wikimedia Commons, step-1,000 checkpoint, same prompt grammar:
+
+### 360° orbit — one input, all 12 azimuths
+
+Twelve separate generations (one per 30° azimuth step, eye-level), looped — the camera circles the
+subject while the cat, cushion and scene hold together:
+
+![360 orbit of a fluffy cat across 12 azimuths](examples/orbit_cat.gif)
+
+### Individual pose comparisons
+
+**Golden retriever · `<mva> front-right quarter view, eye-level shot`**
+![dog quarter](examples/real_0.jpg)
+
+**Golden retriever · `<mva> back view, eye-level shot`**
+![dog back](examples/real_1.jpg)
+
+**Cat · `<mva> front-right quarter view, eye-level shot`**
+![cat quarter](examples/real_2.jpg)
+
+**Cat · `<mva> back view, eye-level shot`**
+![cat back](examples/real_3.jpg)
+
+**Apple · `<mva> front-right quarter view, eye-level shot`**
+![apple quarter](examples/real_4.jpg)
+
+**Apple · `<mva> back view, eye-level shot`**
+![apple back](examples/real_5.jpg)
+
+Generalizes to real photos: the subject rotates to the requested view and the scene mostly survives,
+though identity fidelity on out-of-distribution subjects is good-not-perfect (fine fur detail softens).
+Photo credits (CC): [Golden Retriever Carlos](https://commons.wikimedia.org/wiki/File:Golden_Retriever_Carlos_(10581910556).jpg),
+[Cat03](https://commons.wikimedia.org/wiki/File:Cat03.jpg), [Red Apple](https://commons.wikimedia.org/wiki/File:Red_Apple.jpg).
+
+## Object examples (training distribution)
+
+Front view in, captioned angle out — rendered at the recommended step-1,000 checkpoint:
+
+**`<mva> right side view, high-angle shot`**
+![right side, high-angle](examples/example_1.jpg)
+
+**`<mva> back view, eye-level shot`** — the front features are gone; genuinely the back of the object
+![back view](examples/example_2.jpg)
+
+**`<mva> front-left quarter view, eye-level shot close-up`**
+![front-left close-up](examples/example_3.jpg)
+
+**`<mva> left side view, eye-level shot`**
+![left side](examples/example_4.jpg)
+
+### What the base model does with the same captions
+
+These are step-0 renders from training — the adapter is zero-initialized, so they show the
+**base Qwen-Image-2.1** attempting the identical caption + control image: the camera mostly stays on
+the original front/three-quarter vantage in both natural-language and `<mva>` phrasings.
+
+**base · `<mva> right side view, high-angle shot`**
+![base attempt 1](samples/qwen21_multiple_angles_smoke/samples/1791330024617__000000000_0.jpg)
+
+**base · `<mva> back view, eye-level shot`**
+![base attempt 2](samples/qwen21_multiple_angles_smoke/samples/1791330024617__000000000_1.jpg)
+
+**base · `<mva> front-left quarter view, eye-level shot close-up`**
+![base attempt 3](samples/qwen21_multiple_angles_smoke/samples/1791330051188__000000000_2.jpg)
+
+All 40 sample grids across training (4 poses × every 250 steps, fixed prompts) are in
+[`samples/`](samples/) — including the step-2,500 grid if you prefer the stronger-baked look.
+
+## Prompt format
+
+```
+<mva> {azimuth}, {elevation}[ close-up]
+```
+
+### Azimuth (12 steps of 30°)
+
+| Label | Azimuth |
+|---|---|
+| `front view` | 0° |
+| `front-right view` | 30° |
+| `front-right quarter view` | 60° |
+| `right side view` | 90° |
+| `back-right quarter view` | 120° |
+| `back-right view` | 150° |
+| `back view` | 180° |
+| `back-left view` | 210° |
+| `back-left quarter view` | 240° |
+| `left side view` | 270° |
+| `front-left quarter view` | 300° |
+| `front-left view` | 330° |
+
+### Elevation
+
+| Label | Elevation |
+|---|---|
+| `eye-level shot` | 0° |
+| `elevated shot` | 30° |
+| `high-angle shot` | 60° |
+| `top-down shot` | 90° |
+
+Append ` close-up` for the 62%-crop framing. Example:
+
+```
+<mva> right side view, high-angle shot close-up
+```
+
+## Usage
+
+**Gradio Workflow Space** — [akhaliq/qwen21-multiple-angles-workflow](https://huggingface.co/spaces/akhaliq/qwen21-multiple-angles-workflow):
+canvas UI, photo in → camera knobs out.
+
+**ComfyUI** — load the LoRA with any Qwen-Image-2.1 workflow (LoRA strength 0.8–1.0), pass your
+photo as the edit/reference image, prompt with the format above.
+
+**diffusers** (`QwenImage21Pipeline`, edit mode = pass the image):
+
+```python
+image = pipe(
+    prompt="<mva> back view, eye-level shot",
+    image=input_image,          # front view of your subject
+    num_inference_steps=40,
+    true_cfg_scale=1.0,
+).images[0]
+```
+
+## Training
+
+- **Data**: 5,028 angle-labeled edit pairs from
+  [zeyuanyin/Dome-Objaverse](https://huggingface.co/datasets/zeyuanyin/Dome-Objaverse) (CC-BY-4.0)
+  — 2,514 objects, front view → 2 target views each, composited on a studio background.
+  Published: [akhaliq/qwen21-multiple-angles-train-v1](https://huggingface.co/datasets/akhaliq/qwen21-multiple-angles-train-v1).
+- **Trainer**: ai-toolkit, `arch: qwen_image_2`, control-image conditioning (block-causal reference
+  attention), rank 32 / alpha 32, adamw8bit, lr 1e-4, weighted timesteps, fp8 (DiT + TE), caption
+  dropout 0.05, 512², 2,500 steps.
+- **Checkpoints**: all ten 250-step saves + samples are in `checkpoints/` and `samples/` —
+  step 1,000 recommended, 1,500 alternate, 2,500 stronger-baked.
+
+## Scope & limits
+
+- Trained on Objaverse objects; real photos generalize (see Real photos above) but identity
+  fidelity on fur/fine detail is good-not-perfect.
+- Trained at 512²; larger canvases inherit the style but were not trained on.
+- Angles interpolate reasonably between the trained 30° steps; exact azimuth/elevation numbers in
+  prompts are not parsed — use the labels above.
+
+Credit to [fal's Multiple-Angles LoRA for Edit-2511](https://huggingface.co/fal/Qwen-Image-Edit-2511-Multiple-Angles-LoRA)
+for the caption grammar this format standardized on.

checkpoints/ab_qwen21_angles_ab_base_qwen21_angles_ab_base.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b728f8ae657050dec75406cf85b7f807d52e45750cee6b1a1660fa7b53bdf4a3
+size 159435960

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:945219ec9c7167a6db338c490674bc3747f8802039e4742e433cbaf3a7c9ffbb
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000000250.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fecc88aa95a5f5d4e0bacc9a2c136209abb68d7b459dab3624f8c3c43388a9ec
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000000500.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fdd7d17fc324f47ce99927fdd47d782bd3524fa0dcd453060d49dfacd6f5a9b
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000000750.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6aad0b403e2fa4596b2080a94b1f7d91299fddc21e0a5c9bf11955d939e2a34
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c529c1652a2e5719469a64d8468ed1758c0fe294c66110d8000563f5901a7ea8
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001250.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43d58a50779febfe4703be3c1beabfed090ec17407871ffc208d44920a714ca0
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001500.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:612d52230b0906069cbf108dfbba1e787087c03dd80aadd0f00e9b14192d1f39
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001750.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9f4d2cf01817fb6b3029027a69b8d8f54275582ce3d3347ac33e06376be6f89
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000002000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb3cc3f0ce5b3e7626794b5f69381f651b378afe400161ec7e83936c8361d8e1
+size 159435968

checkpoints/angles_full_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000002250.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:694221b867155e189b8c05a5e5b823772e6a95048182693d40067ead5c1283bc
+size 159435968

checkpoints/angles_smoke_qwen21_multiple_angles_smoke_qwen21_multiple_angles_smoke.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3781a67b1e7153f0e6d91c1451bc2b7b02eb1ca7f744540e2622610643961ff9
+size 159435968

checkpoints/base_real_qwen21_angles_base_real_qwen21_angles_base_real.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f644c49771f78660a215e2f94deb688b1afd5639a046aaf60a769c809dfbff1c
+size 159435968

checkpoints/bench_lora_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e706c38f8e232ab51edd460afd6fbc8f05004a8744f68dca57103b7622ddf13f
+size 159435968

checkpoints/bench_lora_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c529c1652a2e5719469a64d8468ed1758c0fe294c66110d8000563f5901a7ea8
+size 159435968

checkpoints/realphotos2_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a526e13b3111a8668e670e31e7724335070e76f329bee83ba94f48fadc73548
+size 159435968

checkpoints/realphotos2_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c529c1652a2e5719469a64d8468ed1758c0fe294c66110d8000563f5901a7ea8
+size 159435968

checkpoints/realphotos_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10150d4a4042b6c6fd39c50f8964f25c724824ffd6c6dc9855e689ad4abc4810
+size 159435968

checkpoints/realphotos_qwen21_multiple_angles_v1_qwen21_multiple_angles_v1_000001000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c529c1652a2e5719469a64d8468ed1758c0fe294c66110d8000563f5901a7ea8
+size 159435968