fruit detection mainly based on faster rcnn model in pytorch
Original outputs | Outputs with group label |
---|---|
The faster rcnn uses a resnet50 backbone.You can change the depth of the resnet model. Depth must be one of 18, 34, 50, 101 or 152. Note that deeper models are more accurate but are slower and use more memory.
edge guided image outpainting
origin input | cropped_input | edge_input | result_without_edge | result_with_edge |
---|---|---|---|---|
input | output |
---|---|
a giraffe standing next to a tree in a field | |
a bunch of food is on a table | |
a table with a vase of flowers on the table |