garryling/depth_anything_v2_optimized

By garryling · ↓ 0 · ♥ 5 · Updated 2025-04-04

What's in this repo

This page lists the files in garryling/depth_anything_v2_optimized, including depth_anything_v2.onnx, depth_anything_v2_mha.onnx.

2 ONNX models (.onnx files)
5 files, including 2 .onnx, 1 .md, 1 .pt
Last updated 2025-04-04

ONNX files

File	Size	Download
depth_anything_v2.onnx	—	Download
depth_anything_v2_mha.onnx	—	Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

🔍 Depth Anything V2 - Ultra Fast

State-of-the-Art Depth Estimation Meets Mobile Efficiency
Optimized for Snapdragon NPUs with 2X Speed Boost

⚡ Benchmark Comparison

Input resolution:518x518 Snapdragon 8 Gen2 (NPU Acceleration) |Metric | depth_anything_v2.onnx | depth_anything_v2_mha.onnx | Improvement| |-------|--------|-------------|--------| |Latency | 152ms | 73ms | 2.08×| |Memory | 102MB | 45MB | 2.26×|

benchmark method

please intall qai_hub first

import qai_hub as hub

compile_job = hub.submit_compile_job(
    model="depth_anything_v2.onnx",
    device=hub.Device("Samsung Galaxy S23 (Family)"),
)
assert isinstance(compile_job, hub.CompileJob)

profile_job = hub.submit_profile_job(
    model=compile_job.get_target_model(),
    device=hub.Device("Samsung Galaxy S23 (Family)"),
)

reference

Model info

depth-estimation

onnx depth-estimation mobile-optimized transformer qualcomm region:us

↓ 0 ♥ 5

Created: 2025-04-04

Updated: 2025-04-04

View on Hugging Face

Files (5)

.gitattributes

README.md

depth_anything_v2.onnx ONNX

depth_anything_v2_mha.onnx ONNX

depth_anything_v2_mha.pt

Related models

More from garryling

No other models by this author.

More depth-estimation models

Browse all depth-estimation models