This repository contains the code and implementation details of the CascadeTabNet paper "CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents"
MIT License
1.5k
stars
430
forks
source link
Upgraded 2.x model gives different bbox output than 0.4.3 model #181
Hi, I have used upgrade_model_version.py to upgrade my 0.4.3 mmcv version model to 2.1.0 and used it's corresponding config file (cascade-mask-rcnn_hrnetv2p-w32_20e_coco_version2.py) to get the inference. cascade-mask-rcnn_hrnetv2p-w32_20e_coco_version2.txt
My old model provides below output:
bbox_old = [[156.74017, 317.23904, 1056.9825, 880.92944]]
and upgraded model provides below output:
bbox_new = [[182.91272, 343.5786 , 1081.461 , 908.35626]]]
Plotted image: (Green bbox = old model, Blue bbox = upgraded model)
I have noticed that for all of the image inferences the output is slightly shifted towards right and slightly lower.
Please guide me on how to rectify this.
Hi, I have used upgrade_model_version.py to upgrade my 0.4.3 mmcv version model to 2.1.0 and used it's corresponding config file (cascade-mask-rcnn_hrnetv2p-w32_20e_coco_version2.py) to get the inference. cascade-mask-rcnn_hrnetv2p-w32_20e_coco_version2.txt My old model provides below output: bbox_old = [[156.74017, 317.23904, 1056.9825, 880.92944]]
and upgraded model provides below output: bbox_new = [[182.91272, 343.5786 , 1081.461 , 908.35626]]]
Plotted image: (Green bbox = old model, Blue bbox = upgraded model)
I have noticed that for all of the image inferences the output is slightly shifted towards right and slightly lower. Please guide me on how to rectify this.