no unverified claims, CC0 data section, embedded examples

README.md

@@ -16,9 +16,10 @@ pipeline_tag: image-to-image
 
 **Hybrid Efficient Attention with Rank-factorized bias Transformer**
 
-A super-resolution architecture that is simple to train, fast to run, and
-works everywhere: **traiNNer-redux**, **chaiNNer**, **ONNX Runtime**,
-**TensorRT**.
+A super-resolution architecture that is simple to train and fast to run.
+Released models are **verified** with **traiNNer-redux** (training) and
+**ONNX Runtime** (inference); TensorRT and other runtimes are expected to work
+(same standard op set) but are not tested here.
 
 Part of the **BODY** suite by **Philip Hofmann** — networks built to be
 real-world usable first, not benchmark-chasing. See
@@ -27,6 +28,12 @@ real-world usable first, not benchmark-chasing. See
 > Short version: copy `heart_arch.py` into traiNNer-redux, train with
 > `lq_size: 96`, export to ONNX, use anywhere. Details below.
 
+![HEART 4x on anime — input | bicubic | release | OTF GAN](examples/ani_compare.png)
+*Input (as-is) | bicubic x4 | HEART 4x release | HEART 4x OTF GAN.*
+
+![HEART 4x on a noisy image](examples/noisy_compare.png)
+![HEART 4x on a Stable-Diffusion image](examples/stablediffusion_compare.png)
+
 ---
 
 ## Quickstart
@@ -65,8 +72,11 @@ your training config (e.g. `models/heart_4x_pretrain.safetensors` for a 4x run).
 
 ### As a user (want to upscale images)
 
-Load any `.safetensors` model in **chaiNNer** (spandrel auto-detects it), or
-run the `.onnx` files with ONNX Runtime / TensorRT.
+- **ONNX Runtime** (verified): run the `onnx/*.onnx` files (dynamic H/W).
+  TensorRT should also work (same ops) but is untested here.
+- **chaiNNer**: load the `onnx/*.onnx` files directly. Loading the
+  `.safetensors` via spandrel needs HEART registered in your chaiNNer/spandrel
+  build (not upstream yet).
 
 - Input: RGB, any size (dynamic H/W).
 - Scale: 4x (`heart_4x_*`), 2x (`heart_2x`).
@@ -164,6 +174,16 @@ in [`docs/EXPERIMENTS.md`](docs/EXPERIMENTS.md). Highlights:
 - **OTF GAN lq96 window alignment**: diagnosed the bright band + tile grid and
   fixed it via the `lq_size` multiple-of-32 rule.
 
+## Training data & license
+
+All released HEART models were trained **only** on
+[`Phips/lucid-cc0-v2-hc-512`](https://huggingface.co/datasets/Phips/lucid-cc0-v2-hc-512),
+a **CC0** dataset — the released weights are clean to use commercially.
+
+HEART ties HAT on quality at a much lower cost, but we did **not** chase the
+benchmark leaderboard. If higher PSNR/LPIPS is your goal, training on larger
+academic datasets (DIV2K, Flickr2K, LSDIR, ...) is expected to help.
+
 ## Relation to NERVE
 
 HEART and NERVE are the two members of the **BODY** suite (by Philip Hofmann):