Video Quality Assessment
Video Quality Assessment is a computer vision task aiming to mimic video-based human subjective perception. The goal is to produce a mos score, where higher score indicates better perceptual quality. Some well-known benchmarks for this task are KoNViD-1k, LIVE-VQC, YouTube-UGC and LSVQ. SROCC/PLCC/RMSE are usually used to evaluate the performance of different models.
Papers
Showing 1–10 of 216 papers
All datasetsMSU SR-QA DatasetKoNViD-1kMSU NR VQA DatabaseLIVE-VQCMSU FR VQA DatabaseYouTube-UGCLIVE-FB LSVQLIVE-ETRILIVE LivestreamLIVE-YT-HFR
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | DOVER (end-to-end) | PLCC | 0.87 | — | Unverified |
| 2 | ReLaX-VQA (finetuned on YouTube-UGC) | PLCC | 0.87 | — | Unverified |
| 3 | DOVER (head-only) | PLCC | 0.86 | — | Unverified |
| 4 | FasterVQA (fine-tuned) | PLCC | 0.86 | — | Unverified |
| 5 | SimpleVQA | PLCC | 0.86 | — | Unverified |
| 6 | FAST-VQA (finetuned on YouTube-UGC) | PLCC | 0.85 | — | Unverified |
| 7 | HVS-5M | PLCC | 0.85 | — | Unverified |
| 8 | ReLaX-VQA (trained on LSVQ only) | PLCC | 0.84 | — | Unverified |
| 9 | CONVIQT | PLCC | 0.82 | — | Unverified |
| 10 | ReLaX-VQA | PLCC | 0.82 | — | Unverified |