Mean Shift
A clustering method where each point climbs toward the nearest density peak, cluster count and shape both falling out of the data instead of being chosen.
Every point moves repeatedly toward the weighted mean of its neighborhood within a radius called the bandwidth, climbing local density until it settles at a mode — a peak where surrounding density stops increasing. Points converging to the same mode form one cluster.
Unlike k-means, no cluster count is chosen in advance, and uneven cluster sizes are handled naturally, since climbing is a local decision rather than a comparison against every other cluster.