Instructions to use Xalcker/perspectia-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LiteRT
How to use Xalcker/perspectia-models with LiteRT:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Big-LaMa for on-device inpainting (Core ML and LiteRT)
Conversions of Big-LaMa (advimman/lama, Apache-2.0) for running on phones, used by the PerspectIA app to fill the background hidden behind objects in its 3D view. The weights are the original Big-LaMa weights (trained on Places2), unchanged; only the format and the way the Fourier transform is computed differ.
Files
| File | Platform | Size | Notes |
|---|---|---|---|
coreml/lama_big_512_p8.mlpackage |
iOS 17+ | 50 MB | mlprogram, float16, weights palettized to 8 bits. Run it with computeUnits = .cpuAndGPU (the FFT is not supported on the Neural Engine). |
tflite/lama_big_dft_512_fp16.zip |
Android (LiteRT) | 95 MB (104 MB unzipped) | Recommended. Same as below with float16 weights (each followed by a DEQUANTIZE); inputs, output and CPU math stay float32. On a float16 NPU the result is identical. |
tflite/lama_big_dft_512.zip |
Android (LiteRT) | 190 MB (206 MB unzipped) | float32, FFT replaced by matrix products (DFT) so the whole graph runs on GPU/NPU delegates. NHWC inputs/output. |
Inputs and output (512Γ512):
- Core ML:
image(1, 3, 512, 512) RGB in [0, 1];mask(1, 1, 512, 512), 1 = fill. Outputfilled, same shape asimage: the image with the masked area filled in. - TFLite:
image(1, 512, 512, 3) andmask(1, 512, 512, 1), same meaning; output (1, 512, 512, 3).
Conversion
- Source: advimman/lama at commit
7dee0e4a3cf5f73f86a820674bf471454f52b74f; Big-LaMa weights from the smartywu/big-lama copy, revision05cb2be7f8dbe6ca7c6e78f4fc827a4b2baaa4a9(big-lama.zipSHA-256f1b358ca24093b93a106183b98a3dea6e8ed09f3b43ea7251eb2c81e7b4575f6). - The checkpoint is loaded with
torch.load(weights_only=True); no code from the file is executed. - Wrapper as in Qualcomm AI Hub's LaMa: the masked area is blanked out, the mask is passed as a 4th channel, and the prediction is pasted back only inside the mask.
- Core ML:
torch.jit.traceβ coremltools 8.3, float16, then k-means palettization to 8 bits. - TFLite: each
FourierUnit'srfftn/irfftn(normortho) is replaced by an equivalent DFT written as matrix products, all in 4-D tensors; the threeConvTranspose2dwithoutput_paddingare rewritten as an unpadded transposed convolution plus a crop. Exported with litert-torch 0.10 (ai-edge-torch 0.7.2). The float16 variant casts the weights of CONV_2D, TRANSPOSE_CONV and FULLY_CONNECTED with ai-edge-quantizer (FLOAT_CASTING, weight-only).
Accuracy and speed
Against the original PyTorch model (with FFT), inside the mask:
| Mean abs. error | Speed | |
|---|---|---|
| Core ML, 8 bits | 0.050 (palettization) | ~250 ms on a Mac GPU (Apple Silicon) |
| TFLite, float32 | 3e-6 | 184 ms on a Snapdragon 8 Elite NPU (QNN HTP, float16) |
| TFLite, float16 weights | 2.4e-3 on CPU; on the NPU, identical to float32 | 166 ms on the same NPU |
On the Snapdragon NPU, QNN compiles the graph the first time (2β4 minutes); with QNN's context cache later loads take under a second.
License and attribution
Apache-2.0, like the original repository (see LICENSE). LaMa: Suvorov et al., Resolution-robust Large Mask
Inpainting with Fourier Convolutions, WACV 2022. The weights were trained by the LaMa authors on the Places2 dataset;
check that dataset's terms for your use case.
- Downloads last month
- 5