search

karfly / learnable-triangulation-pytorch

This repository is an official PyTorch implementation of the paper "Learnable Triangulation of Human Pose" (ICCV 2019, oral). Proposed method archives state-of-the-art results in multi-view 3D human pose estimation!
MIT License
1.09k stars 181 forks source link

Mask-RCNN and SSD bounding boxes on H36M #159

Closed Billccx closed 1 year ago

Billccx commented 1 year ago

Hi! Thanks for your research contribution. I want to use Mask-RCNN or SSD bounding boxes instead of H36M's ground truth. Would you please share the following files?

detections_paths = {
      'MRCNN': {
          'train': "/Vol1/dbstore/datasets/Human3.6M/extra/train_human36m_MRCNN.json",
          'test': "/Vol1/dbstore/datasets/Human3.6M/extra/test_human36m_MRCNN.json"
      },
      'SSD': {
          'train': "/Vol1/dbstore/datasets/k.iskakov/share/ssd-detections-train-human36m.json",
          'test': "/Vol1/dbstore/datasets/k.iskakov/share/ssd-detections-human36m.json"
      }
  }

I am looking forward to your reply and thanks in advance!

karfly commented 1 year ago

Unfortunately, these files are lost. We are unable to recover them, as we no longer work at Samsung