shadow-cann/hispark-modelzoo-depth-anything-v2

Depth-Anything-v2

Depth Anything V2在细节和鲁棒性方面显著优于 V1。与基于 SD 的模型相比，它具有更快的推理速度、更少的参数和更高的深度精度。本示例使用的是Depth-Anything-V2-Small。

Mirror Metadata

• —Hugging Face repo: shadow-cann/hispark-modelzoo-depth-anything-v2
• —Portal model id: ktgi9vl8j400
• —Created at: 2026-09-12 10:52:18
• —Updated at: 2026-09-15 15:27:10
• —Category: 计算机视觉

Framework

• —PyTorch

Supported OS

• —OpenHarmony
• —Linux

Computing Power

• —Hi3403V100 SVP_NNN
• —Hi3403V100 NNN

Tags

• —单目深度

Detail Parameters

• —输入: 518x518
• —参数量: 24.184M
• —计算量: 127.785GFLOPs

Files In This Repo

• —depth_anything.om (编译模型 / A16W8 / A16W8)
• —depthanything.om (编译模型 / FP16 / FP16)
• —depthanythingv2_vits.pth (源模型 / 源模型下载)
• —depthanythingv2vitssvp_nnn.onnx (源模型 / 源模型下载)
• —depthanythingv2vitsnnn.onnx (源模型 / 源模型下载)

Upstream Links

• —Portal card: https://gitbubble.github.io/hisilicon-developer-portal-mirror/model-detail.html?id=ktgi9vl8j400
• —Upstream repository: https://gitcode.com/HiSpark/modelzoo/tree/master/samples/built-in/depth/Depth-Anything-v2
• —License reference: https://github.com/DepthAnything/Depth-Anything-V2/blob/main/LICENSE

Notes

• —This repository was mirrored from the HiSilicon Developer Portal model card and local downloads captured on 2026-03-27.
• —File ownership follows the portal card mapping, not just filename similarity.
• —Cover image: 1700942867267585_depth.png