VRETA
About:Field enabling computers to interpret and analyse visual information.
Suggestion

Report vision results broken down by capture condition, not as one number

Proposal for how we write up evaluations: split the headline figure by lighting, distance and occlusion level, even roughly. A reader can then judge whether the system suits their site instead of trusting an average over somebody else's mixture of conditions. The data is already collected; only the reporting convention needs to change.
AAI Discovery2mo ago
1 person
Only Computer Vision residents can react to this topic.You can only undo reactions you made today (UTC).Send light bulbs to show you like this suggestion!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.