search

chufengt / ALM-pedestrian-attribute

Code for the paper "Improving Pedestrian Attribute Recognition With Weakly-Supervised Multi-Scale Attribute-Specific Localization", ICCV 2019, http://arxiv.org/abs/1910.04562.
Apache License 2.0
188 stars 66 forks source link

Differentiate ROI pooling implementation #53

Closed abhigoku10 closed 3 years ago

abhigoku10 commented 3 years ago

@chufengt can you share the source code for the architecture having differentiate ROI pooling or the differentiate ROI pooling module wanted to check the accuracy and conversion to onnx for PAR, thanks in advance

chufengt commented 3 years ago

Sorry, I just said we need an operation similar to a differentiable ROI pooling module, and we choose STN in the end. We didn't implement any model with an explicit differentiable ROI pooling module.

abhigoku10 commented 3 years ago

@chufengt oh wokay since STN we are not able to convert to tensorRT so thought we shall replace it with ROI pooling , is there any other modules which i can replace STN with ??

chufengt commented 3 years ago

maybe a box predictor + RoI pooling the predictor used to predict the cropped region in STN