2017-10-02 Lossy Compression WG Meeting notes

Date

02 Oct 2017

Attendees

• Robert Gruendl

Topics

1. Q/A with Kyzystof (in case there are any areas where we can bring them up to "speed")
2. Further discussion about test inputs
3. Discuss the initial mode of testing (and try to outline a first set of tests)
   • compress/uncompress speeds
   • compression factor
   • image/catalog comparison (before and after compression)

Notes:

• BlueJeans connection precluded meeting but the only other attendee was Paul Price
• After ~20 minutes of messing with BlueJeans connections switched to Skype.
• Discussion with Paul led to cutting back on weekly cadence of meetings.
• Also, Paul indicates that the framework to test compression modes already exists within package developed for HSC data.
  • the implementation current has a modified version of the FPACK/RICE algorithm that eliminates objects when estimating noise (for the quantization) 
    
    

Action items