| { | |
| "architectures": [ | |
| "VideoDepthAnythingForDepthEstimation" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_video_depth_anything.VideoDepthAnythingConfig", | |
| "AutoModel": "modeling_video_depth_anything.VideoDepthAnythingForDepthEstimation", | |
| "AutoModelForDepthEstimation": "modeling_video_depth_anything.VideoDepthAnythingForDepthEstimation" | |
| }, | |
| "dtype": "float32", | |
| "encoder": "vitl", | |
| "features": 256, | |
| "input_size": 518, | |
| "metric": false, | |
| "model_type": "video_depth_anything", | |
| "num_frames": 32, | |
| "out_channels": [ | |
| 256, | |
| 512, | |
| 1024, | |
| 1024 | |
| ], | |
| "pe": "ape", | |
| "transformers_version": "5.20.2", | |
| "use_bn": false, | |
| "use_clstoken": false | |
| } | |