Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Learning Video Dynamics with Predictive Differentiable Rendering

About

How to accurately predict a high-fidelity future world? While the visual world is inherently continuous, existing deterministic video prediction models operate in discrete pixel space and are mainly optimized with pixel-wise mean squared error (MSE), which often leads to over-smoothed predictions and a lack of fine-grained visual details. To address these limitations, we propose Predictive Differentiable Rendering (PDR), a novel end-to-end video prediction paradigm that bridges the gap between discrete and continuous representations. Inspired by recent progress in 3D reconstruction with 3D Gaussian Splatting, we introduce PredGS, a lightweight and plug-and-play adapter based on 2D Gaussian representation, which could be seamlessly integrated with existing pixel space predictors, significantly improving spatial detail preservation with negligible computational overhead. Furthermore, we develop predgsplat, a CUDA-accelerated differentiable 2D Gaussian renderer supporting arbitrary channels. Each Gaussian is defined by 5 + C learnable parameters (position, scale, rotation, and C channel amplitudes) and achieves up to 10x faster rendering than the baseline. Optimized by a combined L1 and SSIM loss, PDR overcomes the inherent blurring tendencies of MSE Loss, significantly enhancing the prediction performance. Extensive experiments on diverse real-world benchmarks, including TaxiBJ, WeatherBench, KTH, and Human3.6M, demonstrate that PDR consistently surpasses existing methods, delivering superior detail preservation, visual fidelity, and predictive accuracy.

Yujin Tang, Tian Zhou, Xin Lin, Cheng Tan, Yifan Hu, Rong Jin, SouYoung Jin, Liang Sun• 2026

Related benchmarks

TaskDatasetResultRank
Video PredictionKTH 10 -> 20 steps (test)
PSNR26.98
102
Video PredictionHuman3.6M 4 → 4 frames (test)
SSIM98.57
16
Video PredictionTaxiBJ 4 → 4 frames (test)
MSE0.2807
15
Video PredictionWeatherBench T2m 12 -> 12 frames (test)
MSE1.071
14
Video PredictionKTH 64x64, 10 → 30 frames
PSNR26.69
7
Showing 5 of 5 rows

Other info

Follow for update