Convert depth maps into high-quality images
Fonte do modelo
Descrição da fonte
Depth LoRA
Convert depth maps into high-quality images
Introduction
This is a depth LoRA that converts depth maps into high-quality images. It was trained in multiple stages on a variety of datasets to balance overall structural accuracy with fine-grained detail reconstruction.
HF download:
https://huggingface.co/lrzjason/Krea2-Lora-Collection/tree/main/loras
EditUtils
Fontes
1 fonteArtefatos de modelo
1 artefatokrea2_depth_lrzjason.safetensors
safetensors · BF16 · 872 MB · SHA-256 dfa2051e7ac7…ff36 · Civitai
BaixarTrechos de fonte
1 trechoSource context: 754 downloads · Type LORA · Base model Krea 2
Training Details
The training followed a two-stage, progressive-resolution strategy:
Stage 1 — 512 Resolution Pretraining
Trained on a 2,000-pair subset sampled from a public depth dataset, with all captions regenerated by a VLM (VLM re-caption) to improve text-image alignment. This stage used a constant learning rate of 1e-4 to establish a solid depth-structure prior.
Stage 2 — 1536 High-Resolution Refinement
Continued training on a 1536-resolution depth dataset, focusing on detail reconstruction (edges, thin structures, and fine geometry). This stage switched to a cosine learning rate schedule for more stable convergence.
Stage 1 512 res, 2,000 pairs (public depth dataset + VLM re-caption), constant LR 1e-4 Stage 2 1536 res depth dataset, cosine LR scheduleTotal Epochs50+
Usage
⚠️ This LoRA must be used together with the EditUtils plugin: 👉 https://github.com/lrzjason/ComfyUI-EditUtils
Install the ComfyUI-EditUtils plugin.
Load this LoRA in your workflow.
Feed in the source image to obtain the corresponding depth map.
Notes
Works better on 10240- 2048 with long captions; with higher-resolution inputs yielding richer depth details.
Will not work as intended without the EditUtils plugin.
RH Workflow
📬 Contact
X Twitter : @Lrzjason
📧 Email : [email protected]
💬 QQ Group : 866612947
💬 WeChat ID : fkdeai
🎨 CivitAI : xiaozhijason