I am trying to use the convolutional-pose-machines to find the pose of people in an office environment, where people are usually seated. My question is; is there a caffemodel that finds the upper body parts of a person without their lower body (when people are sitting the legs are usually occluded)? The first part of the code (person_net) is failing to find seated people.
You're right, the person detector here is not very strong. You can replace it to other person detector and use to result as the initialization of CPMs.
Hi,
I am trying to use the convolutional-pose-machines to find the pose of people in an office environment, where people are usually seated. My question is; is there a caffemodel that finds the upper body parts of a person without their lower body (when people are sitting the legs are usually occluded)? The first part of the code (person_net) is failing to find seated people.
Help is much appreciated