VisDet Model Weights

This repository contains pretrained weights for VisDet, a streamlined object detection and instance segmentation library.

Models

The weights are organized by source:

  • openmmlab/: Backbone weights from OpenMMLab's model zoo
  • mmcls/: Classification backbone weights from MMClassification

Usage

from visdet.apis import init_detector

# Use HuggingFace-hosted weights
model = init_detector(
    config="configs/mask_rcnn/mask_rcnn_r50_fpn_1x_coco.py",
    checkpoint="hf://GeorgePearse/visdet-weights/openmmlab/resnet50_msra-5891d200.pth"
)

Or in config files:

model = dict(
    backbone=dict(
        init_cfg=dict(
            type="Pretrained",
            checkpoint="hf://GeorgePearse/visdet-weights/openmmlab/resnet50_msra-5891d200.pth"
        )
    )
)

License

These weights are provided under the Apache 2.0 license, consistent with their original sources.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support