eess.IVNov 16, 2026

How Much Does Saliency Accuracy Matter for Saliency-Guided 360-degree Image Compression?

Authors: Sakib Ahmed, Sidi-Mohammed Senouci, Philippe Brunet, Guillaume Caron, Nevrez Imamoglu

Organizations: Université Bourgogne Europe, DRIVE UR 1859, 58000 Nevers, France · MIS laboratory, University of Picardie Jules Verne, Amiens, France · CNRS-AIST JRL, AIST, Tsukuba, Japan · Artificial Intelligence Research Center, AIST, Tsukuba, Japan

Abstract

Saliency-guided coding allocates bits to the regions viewers are most likely to attend, but accurate 360-degree saliency prediction is computationally expensive, an obstacle for low-latency immersive streaming applications such as robot teleoperation. We quantify how much saliency-prediction accuracy actually translates into compression gain. Building on a saliency-guided rate-distortion-optimized (RDO) learned codec, modified so that the training-time latent mask is driven by ground-truth saliency, which yields a single, predictor-agnostic codec, we evaluate saliency predictors spanning two orders of magnitude in compute on both the learned codec and H.265/HEVC. The results show that higher saliency-prediction accuracy yields diminishing compression gains: a lightweight predictor fast enough for live use (52 frames per second) captures 65% of the ground-truth oracle's bit-rate saving, a 3.3x heavier one reaches 74%, and a further 38x increase in compute adds no measurable benefit, on either codec. Therefore, lightweight predictors might also be a practical choice for latency-constrained 360-degree streaming.

Figures & tables

Explore similar work

CardsList
  1. Viewport-based Neural 360° Image Compression

    Mar 24, 2026Jingwei Liao, Bo Chen, Klara Nahrstedt +1Learned Image Compression

  2. SCOPE: Semantic COdec Pre-Editing for Object-Preserving Compression

    Oct 29, 2026Silvia Dani, Leonardo Galteri, Marco BertiniTask-Aware Compression