Simple Online and Realtime Tracking (SORT) is a pragmatic approach tomultiple object tracking with a focus on simple, effective algorithms. In thispaper, we integrate appearance information to improve the performance of SORT.Due to this extension we are able to track objects through longer periods ofocclusions, effectively reducing the number of identity switches. In spirit ofthe original framework we place much of the computational complexity into anoffline pre-training stage where we learn a deep association metric on alarge-scale person re-identification dataset. During online application, weestablish measurement-to-track associations using nearest neighbor queries invisual appearance space. Experimental evaluation shows that our extensionsreduce the number of identity switches by 45%, achieving overall competitiveperformance at high frame rates.