AI Paper Atlas

Research map / Video / Video segmentation and tracking

MFQE 2.0: A New Approach for Multi-Frame Quality Enhancement on Compressed Video

Zhenyu Guan, Qunliang Xing, Mai Xu, Ren Yang, Tie Liu, Zulin Wang

TPAMI 2019 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 265 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use PQFs as reference instead of neighboring frames (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.274 -> 0.562Improved
use ground-truth PQF labels instead of detected (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.562 -> 0.563No clear effect
add post-processing to PQF detector (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 98.15 -> 98.21No clear effect
replace SVM detector with BiLSTM (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.528 -> 0.562Improved
add dense connections to QE-subnet (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.299 -> 0.562Improved
add multi-scale feature extraction to QE-subnet (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.278 -> 0.299Improved
enlarge training database from 70 to 160 sequences (ablation)MFQE 2.0 · JCT-VC 18 sequences, HEVC QP=37Delta PSNR (dB): 0.533 -> 0.562Improved
retrain on H.264 data instead of HEVC data (ablation)MFQE 2.0 · H.264 QP=37Delta PSNR (dB): 0.422 -> 0.464Improved
replace MFQE 1.0 network with MFQE 2.0 lightweight MF-CNNMFQE · WQXGA test speedfps: 0.36 -> 0.79Improved
replace MFQE 1.0 with MFQE 2.0MFQE · JCT-VC 18 sequences, BD-BRBD-BR reduction (%): 11.41 -> 14.06Improved

Later papers that compared with this method

2 later accepted papers report MFQE 2.0 as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy