graph-rcnn.pytorch
graph-rcnn.pytorch copied to clipboard
[ECCV 2018] Official code for "Graph R-CNN for Scene Graph Generation"
I would like to get a pretrained model of the graph-rcnn and be able to run it on any images that I might choose. I'm unable to understand anything from...
I run the code with `python main.py --config-file configs/faster_rcnn_res101.yaml` command, and using Mini VG dataset. There is an error saying: Traceback (most recent call last): File "main.py", line 127, in...
Thank you for your good work! But I still can’t reach your score when I trained faster-rcnn. All my configurations are consistent with your default configurations. I don't know why....
Thank you for your good work! I was wondering when you could release the code in the Attention GCN section?The version 0.4.0 you released before cannot open this website because...
I thought that calling the command for "Evaluate scene graph generation model" plus --visualize will generate a graph similar to this but I'm unable to locate the output. Does anyone...
First of all, thanks for a great research and code! I was wondering if there is a way to visualize the scene graph obtained from the object detections. I am...
I am running the graph-rcnn network on a custom dataset and I am recieving the following error. ``` 2020-11-05 18:46:14,177 scene_graph_generation.trainer INFO: Start training 2020-11-05 18:46:17,477 scene_graph_generation.trainer INFO: model: scene_parser...
I am training rcnn with the multi-gpu scripts but got the following error: `RuntimeError: boxlists should have same image size, got BoxList(num_boxes=1, image_width=640, image_height=53 7, mode=xyxy), BoxList(num_boxes=9, image_width=537, image_height=640, mode=xyxy)`...
When evaluation 'sgcls' and 'predcls', I encounter this error. What's the problem? 
Hi Jianwei, thank you for the excellent code! When I was trying to visualize the image and predictions, I find the RGB channels of the image swapped. I looked into...