Summer

I dont think that paper describes how the model itself processes the outputs.

Hello! Im Sammy from the Virtual Reality Computer Vision Group! Currently we are doing research into alternative methods for facial tracking inside HMDs and we would like to feature SRanipal in our first ever paper! However after obtaining the Vive Facial Tracker AI model (The raw model without SRanipal), we quickly realized that the output of the model is not consistent with current documentation. Would it be possible to tell us how SRanipal processes the raw output from the Vive Facial Tracker model?