Is calculation of any bias based on the color of the people in front of a police officer or is it based on the total number of people who share that skin color?
Perhaps addressing both sides of this issue simultaneously will afford us some clarity.
First, we know black people are far more likely to attract police attention than white people are for identical behavior:
http://faculty.chicagobooth.edu/bernd.wittenbrink/research/pdf/cpjw07.pdf
http://medicalxpress.com/news/2016-07-police-people-legal-interventions.html
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2561263/
http://onlinelibrary.wiley.com/doi/10.1111/j.1460-2466.2008.00398.x/abstract
Second, we know police are more likely to shoot if the person is black than if they're white. This has been found both in the research of actual shootings:
http://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0141854
And in the research of simulated confrontations:
http://psych.colorado.edu/~jclab/FPST.html (http://faculty.chicagobooth.edu/bernd.wittenbrink/research/pdf/cpjw07.pdf)
http://pss.sagepub.com/content/16/3/180.abstract
So, from both increased interactions with the police AND a higher error rate to shoot at unarmed black persons, I don't think we should be surprised when we discover that more unarmed black people are shot than unarmed white people. And I don't think we should disparage an entire social movement when they're outraged by this disproportionate use of force.