A fast exhaustive search algorithm for rate-constrained motion estimation

Abstract

A fast exhaustive search algorithm for rate-constrained motion estimation is presented. The motion vectors are selected from a search window based on a rate-distortion criterion by successively eliminating the search positions depending on the rate constraint. The estimation performance of the proposed algorithm is identical to the performance of the rate-constrained full search algorithm, with considerable reduction in computation. Simulation results indicate that the number of matching calculations decreases as the constraint on the rate increases.

An adaptive search ordering for rate-constrained successive elimination algorithms

“…This is analogous to the findings of [9], and is caused by an increase in the value of the Lagrange multiplier. This in turn increases the ratio between the weighted number of bits required to encode the motion vector and the prediction error.…”

“…In [9], Coban and Mersereau added the rate constraint to the successive elimination algorithms. In a rate-constrained video encoder, this constraint must be added to the SEA in order to produce correct results.…”

“…In this paper, we extend the work of [6], [9] to support the H.265/HEVC standard and implement it in the HEVC reference software. However, as described in [12], this leads to unnecessary cost function evaluations.…”

“…Spiral search ordering outperforms raster search ordering by evaluating better block matching candidates earlier in the search process, which in turn filters out more block matching candidates. Spiral search ordering is considered state-ofthe-art, and is used in many implementations of SEA-based algorithms [8,9,11]. However, Trudeau et al [12] showed that a spiral search ordering, in a rate-constrained context, will perform unnecessary cost function evaluations.…”

“…In [9], Coban and Mersereau modified the SEA filtering criterion to take into account the number of bits required to encode the motion vector of a block matching candidate. This modification is in line with the HEVC encoder-side description of the test model reference software [10], where the optimal matching candidate block is the best rate-constrained match.…”