This repository packages the unfold-free ZipDepth Base mobile/NPU checkpoint as a static 384×384 ExecuTorch model. The depth network uses FP16 activations with FP32 serialized convolution weights and external tensors.
ZipDepth predicts affine-invariant inverse depth from one RGB image. The package accepts arbitrary image dimensions up to 4096×4096, crops the largest top-left square, resizes it to the static 384×384 depth-model input, and restores the result to the original tensor shape with zero padding.
a302e543zipdepth_base_npu.pthd13c78971338b35a219e0a27f7045b29c7f4e8c1Follow these steps to run ZipDepth Base FP16 — ExecuTorch on macOS® with Apple Silicon or on Linux® with an Arm®-based CPU, such as a Raspberry Pi® 5.
git clone https://github.com/ARM-software/ComputeFlow.git && cd ComputeFlow
uv sync && source .venv/bin/activate
compute-flow update
compute-flow model install monocular_depth_estimation:zip-depth zipdepth-base-executorch-fp16
compute-flow build native
python -m cli_examples.monocular_depth_estimation_cli \
-m @manifest/zip-depth/zipdepth-base-executorch-fp16/zipdepth_manifest.json \
-i @sample/im0.jpg \
-t 1
The example estimates relative depth for the bundled samples/im0.jpg image. Replace -i with a
path to your own image.
Quality was evaluated on all 654 images in the NYUv2 test split. Both models received the same top-left square crop resized to 384×384. Their predictions were resized to the ground-truth shape, aligned independently in inverse-depth space using a least-squares scale and shift, and evaluated over the NYUv2 Eigen crop and 0.001–10 m depth range. This isolates the static depth models; it is not an evaluation of the dynamic prepare/restore transforms.
| Metric | FP32 | FP16 ExecuTorch | Change |
|---|---|---|---|
| AbsRel ↓ | 15.796% | 15.785% | 0.011 percentage points lower |
| δ1 ↑ | 77.055% | 77.009% | 0.046 percentage points lower |
The FP16 candidate uses a different official ZipDepth checkpoint because the mobile/NPU checkpoint replaces convex unfold upsampling with a deployment-friendly learned blend. The comparison therefore measures the complete mobile graph and FP16 conversion, not FP16 rounding in isolation. On the distributed 384×384 sample, its output has 0.99645 Pearson correlation with the FP32 output.
The static depth models were profiled using the distributed samples/im0.jpg image, which is
already 384×384. The dynamic prepare and restore models were therefore bypassed.
d13c7897 FP16 candidate, both with the
XNNPACK CPU backend.V2502A) on Android™ 16, USB powered at 100% battery.| Metric | Android™ Vivo X300: FP32 | Android™ Vivo X300: FP16 | Uplift |
|---|---|---|---|
| p50 inference latency | 41.987 ms | 20.431 ms | 2.06× faster |
| p90 inference latency | 42.612 ms | 21.342 ms | 2.00× faster |
| p99 inference latency | 43.236 ms | 27.181 ms | 1.59× faster |
| Component role | Framework / format |
|---|---|
| Input square crop and resize | ExecuTorch / PTE |
| ZipDepth Base inference | ExecuTorch / PTE |
| Output resize and zero padding | ExecuTorch / PTE |
| Input | Data type | Shape | Value range |
|---|---|---|---|
| RGB image | FP32 | 3 × H × W, with 1 ≤ H,W ≤ 4096 |
[0, 1] |
| Output | Data type | Shape | Description |
|---|---|---|---|
| Relative depth | FP32 | H × W |
Affine-invariant inverse depth; padded regions are zero |
zipdepth_base_fp16.pte — static FP16 ZipDepth Base mobile/NPU depth model.zipdepth_prepare_image.pte — dynamic input crop-and-resize transform.zipdepth_restore_depth.pte — dynamic output resize-and-padding transform.zipdepth_manifest.json — model tensor contracts and runtime configuration.samples/im0.jpg — 384×384 upstream sample used for profiling.metadata.yaml and benchmarks/ — model and benchmark metadata.LICENSE — upstream MIT license.SHA256SUMS — reproducibility checksums.zipdepth_base_npu.pth
is the source checkpoint for zipdepth_base_fp16.pte.samples/im0.jpg
is an upstream example crop.Verify the checked-out files with:
shasum -a 256 -c SHA256SUMS
@inproceedings{tosi2026zipdepth,
title = {{ZipDepth}: Bringing Lightweight Zero-Shot Monocular Depth Anywhere, on Any Device},
author = {Tosi, Fabio and Bartolomei, Luca and Poggi, Matteo and Mattoccia, Stefano},
booktitle = {European Conference on Computer Vision (ECCV)},
year = {2026}
}