While fishery-dependent data represents the preferential way to collect geographically spreadtime series of data to study the Antarctic toothfish, it is limited to areas where commercial fishing is permitted either by management or vessel operational constraints. The use of Baited Remote Underwater Video (BRUV) allows a solution to this problem. Thanks to the collaboration between New Zealand and Italy, from 2015 to 2018 a series of BRUV samplings have been carried out from the fast ice of McMurdo Sound and Terra Nova Bay, in the southwestern Ross Sea. Here we report a methodological approach to obtain from the videos recorded information on abundance and size distribution related to a series of predictive variables, including an approach to account for bias in counting the number of individuals attracted to the bait. Because we are now focussed on developing the method, the results are preliminary. The aim of the paper is to disseminate to the Members the methodology that we have used. Once refined, our hope is that it could provide a mechanism for other Members to extend the area of data collection and better understand Antarctic toothfish distribution in shelf areas under the ice.