Skip to main content

Table 4 Comparative compression performance of the proposed feedback-free architecture

From: Encoder-driven rate control and mode decision for distributed video coding

   GOP2 GOP4 GOP8
∆R
(%)
∆PSNR
(dB)
∆R
(%)
∆PSNR
(dB)
∆R
(%)
∆PSNR
(dB)
Foreman DISCOVER -4.21 0.21 1.60 -0.13 10.33 -0.56
  [7] -1.80 0.13 -10.42 0.63 -18.36 1.10
  Proposed DVC with feedback -11.55 0.66 -20.43 1.11 -27.94 1.54
Soccer DISCOVER 9.80 -0.56 14.36 -0.84 15.94 -0.89
  [7] -1.28 0.07 -10.66 0.55 -21.55 1.21
  Proposed DVC with feedback -16.04 0.83 -23.82 1.30 -30.68 1.79
Football DISCOVER 2.82 -0.18 1.94 -0.11 1.28 -0.10
  [7] -7.27 0.44 -15.28 0.97 -21.39 1.34
  Proposed DVC with feedback -13.36 0.81 -21.48 1.37 -27.45 1.74
Carphone DISCOVER -4.07 0.21 -9.80 0.47 -5.42 0.23
  [7] -1.57 0.18 -8.49 0.58 -12.27 0.76
  Proposed DVC with feedback -8.94 0.50 -14.28 0.72 -17.79 0.89
  1. Performance is expressed in terms of Bjøntegaard deltas[41] of the proposed feedback-free DVC configured with all three coding modes, versus DISCOVER[13], the hash-based DVC system with feedback from[7] and the proposed DVC system configured with feedback-based decoder-driven rate allocation. Negative rate and positive PSNR delta values correspond to gains compared to the proposed feedback-free solution.