SRT: Super-Resolution for Time Series via Disentangled Rectified Flow
About
Fine-grained time series data with high temporal resolution is critical for accurate analytics across a wide range of applications. However, the acquisition of such data is often limited by cost and feasibility. This problem can be tackled by reconstructing high-resolution signals from low-resolution inputs based on specific priors, known as super-resolution. While extensively studied in computer vision, directly transferring image super-resolution techniques to time series is not trivial. To address this challenge at a fundamental level, we propose Super-Resolution for Time series (SRT), a novel framework that reconstructs temporal patterns lost in low-resolution inputs via disentangled rectified flow. SRT decomposes the input into trend and seasonal components, aligns them to the target resolution using an implicit neural representation, and leverages a novel cross-resolution attention mechanism to guide the generation of high-resolution details. We further introduce SRT-large, a scaled-up version with extensive pre-training, which enables strong zero-shot super-resolution capability. Extensive experiments on nine public datasets demonstrate that SRT and SRT-large consistently outperform existing methods across multiple scale factors, showing both robust performance and the effectiveness of each component in our architecture.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Aggregated Super-Resolution | ETTm1 | MSE2.944 | 10 | |
| Aggregated Super-Resolution | ETTm2 | MSE1.401 | 10 | |
| Aggregated Super-Resolution | ETTh1 | MSE18.897 | 10 | |
| Aggregated Super-Resolution | ETTh2 | MSE8.248 | 10 | |
| Aggregated Super-Resolution | Weather | MSE2.947 | 10 | |
| Aggregated Super-Resolution | PEMS-SF | MSE12.542 | 10 | |
| Aggregated Super-Resolution | SCP1 | MSE7.206 | 10 | |
| Aggregated Super-Resolution | CP2 | MSE8.004 | 10 | |
| Sampled Super-Resolution | ETTm1 (test) | MSE2.539 | 10 | |
| Sampled Super-Resolution | ETTm2 (test) | MSE1.417 | 10 |