Push model using huggingface_hub.

README.md

@@ -0,0 +1,12 @@
+---
+library_name: UniDepth
+tags:
+- monocular-metric-depth-estimation
+- pytorch_model_hub_mixin
+- model_hub_mixin
+repo_url: https://github.com/lpiccinelli-eth/UniDepth
+---
+
+This model has been pushed to the Hub using **UniDepth**:
+- Repo: https://github.com/lpiccinelli-eth/UniDepth
+- Docs: [More Information Needed]

config.json

@@ -0,0 +1,85 @@
+{
+  "data": {
+    "image_shape": [
+      420,
+      560
+    ],
+    "shape_constraints": {
+      "patch_size": 14,
+      "pixels_bounds": [
+        1800,
+        2400
+      ],
+      "ratio_bounds": [
+        0.66,
+        2.0
+      ]
+    }
+  },
+  "eps": 1e-06,
+  "generic": {
+    "deterministic": true,
+    "seed": 13
+  },
+  "model": {
+    "expansion": 4,
+    "name": "UniDepthV2",
+    "num_heads": 8,
+    "pixel_decoder": {
+      "depths": [
+        6,
+        0,
+        0
+      ],
+      "dropout": 0.0,
+      "hidden_dim": 512
+    },
+    "pixel_encoder": {
+      "depths": [
+        21,
+        22,
+        23,
+        24
+      ],
+      "embed_dim": 1024,
+      "embed_dims": [
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024,
+        1024
+      ],
+      "name": "dinov2_vitl14",
+      "output_idx": [
+        21,
+        22,
+        23,
+        24
+      ],
+      "patch_size": 14,
+      "pretrained": null,
+      "stacking_fn": "last",
+      "use_norm": true
+    }
+  },
+  "training": {}
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2aefcdee26bf7f47abb8fb4e2262503acb57d472fe0ce7338b4c0656774e6a3
+size 1451442496