ViSAGE @ NTIRE 2026 Challenge on Video Saliency Prediction
This is an incremental improvement for computer vision researchers and practitioners working on video saliency prediction, as it enhances performance in a specific challenge setting.
The paper tackled the problem of video saliency prediction by proposing ViSAGE, a multi-expert ensemble framework that aggregates diverse inductive biases to capture complex spatio-temporal cues, resulting in a champion solution that ranked first on two out of four metrics and outperformed most competitors on others in the NTIRE 2026 Challenge.
In this report, we present our champion solution for the NTIRE 2026 Challenge on Video Saliency Prediction held in conjunction with CVPR 2026. To exploit complementary inductive biases for video saliency, we propose Video Saliency with Adaptive Gated Experts (ViSAGE), a multi-expert ensemble framework. Each specialized decoder performs adaptive gating and modulation to refine spatio-temporal features. The complementary predictions from different experts are then fused at inference. ViSAGE thereby aggregates diverse inductive biases to capture complex spatio-temporal saliency cues in videos. On the Private Test set, ViSAGE ranked first on two out of four evaluation metrics, and outperformed most competing solutions on the other two metrics, demonstrating its effectiveness and generalization ability. Our code has been released at https://github.com/iLearn-Lab/CVPRW26-ViSAGE.