Research map / Video / Video segmentation and tracking
Video frame interpolation
Synthesizes intermediate frames using transformers, depth-aware warping and motion modeling.
60 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Video segmentation and tracking, keyword cluster “interpolation · flow · event”
State History Benchmarks Tried Baselines Checklist Preprints Who Export
Where it stands
Few recent reports The tracked benchmarks have at most one report from 2025–26 that could beat the old record, too few to say whether they are still moving.
23 papers in 2023–24 and 10 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 5.2% to 1.9%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2017 · Video Frame Interpolation via Adaptive Convolution (CVPR 2017 · 533 citations)
- 2018 · Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (CVPR 2018 · 867 citations)
- 2019 · Depth-Aware Video Frame Interpolation (CVPR 2019 · 577 citations)
- 2020 · Softmax Splatting for Video Frame Interpolation (CVPR 2020 · 395 citations)
- 2021 · CDFI: Compression-Driven Network Design for Frame Interpolation (CVPR 2021 · 99 citations)
- 2022 · IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation (CVPR 2022 · 193 citations)
- 2023 · Extracting Motion and Appearance via Inter-Frame Attention for Efficient Video Frame Interpolation (CVPR 2023 · 135 citations)
- 2024 · Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · 30 citations)
- 2025 · LC-Mamba: Local and Continuous Mamba with Shifted Windows for Frame Interpolation (CVPR 2025 · 8 citations)
- 2026 · VTinker: Guided Flow Upsampling and Texture Mapping for High-Resolution Video Frame Interpolation (AAAI 2026 · 0 citations)
Earliest papers here
- Video Frame Interpolation via Adaptive Convolution (CVPR 2017 · 533 citations)
- Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (CVPR 2018 · 867 citations)
- Context-Aware Synthesis for Video Frame Interpolation (CVPR 2018 · 435 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
UCF101 · PSNR (higher is better)
17 papers, 2018–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2018 | 33.14 | Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation (CVPR 2018 · 867 citations) |
| 2019 | 34.99 | Depth-Aware Video Frame Interpolation (CVPR 2019 · 577 citations) |
| 2020 | 35.08 | Video Frame Interpolation via Deformable Separable Convolution (AAAI 2020 · 145 citations) |
| 2022 | 35.43 | Video Frame Interpolation With Transformer (CVPR 2022 · 143 citations) |
| 2023 | 35.48 | Extracting Motion and Appearance via Inter-Frame Attention for Efficient Video Frame Interpolation (CVPR 2023 · 135 citations) |
Vimeo90K · PSNR (higher is better)
14 papers, 2019–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2019 | 34.71 | Depth-Aware Video Frame Interpolation (CVPR 2019 · 577 citations) |
| 2020 | 34.73 | Video Frame Interpolation via Deformable Separable Convolution (AAAI 2020 · 145 citations) |
| 2021 | 35.17 | CDFI: Compression-Driven Network Design for Frame Interpolation (CVPR 2021 · 99 citations) |
| 2022 | 35.8 | IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation (CVPR 2022 · 193 citations) |
| 2023 | 36.67 | Exploring Discontinuity for Video Frame Interpolation (CVPR 2023 · 6 citations) |
| 2024 | 36.68 | LAVIB: A Large-scale Video Interpolation Benchmark (NeurIPS 2024 · 4 citations) |
SNU-FILM Extreme · PSNR (higher is better)
10 papers, 2020–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2020 | 24.69 | Channel Attention Is All You Need for Video Frame Interpolation (AAAI 2020 · 337 citations) |
| 2022 | 25.81 | ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation (CVPR 2022 · 47 citations) |
SNU-FILM Hard · PSNR (higher is better)
10 papers, 2020–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2020 | 29.86 | Channel Attention Is All You Need for Video Frame Interpolation (AAAI 2020 · 337 citations) |
| 2022 | 31.7 | ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation (CVPR 2022 · 47 citations) |
SNU-FILM Easy · PSNR (higher is better)
7 papers, 2022–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2022 | 40.77 | ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation (CVPR 2022 · 47 citations) |
SNU-FILM Medium · PSNR (higher is better)
7 papers, 2022–2025 · one report since 2024, not better
| Year | Best so far | Set by |
|---|---|---|
| 2022 | 37.11 | ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation (CVPR 2022 · 47 citations) |
What has been tried
58 papers here report 616 measured changes (a component added, removed or swapped, compared with a baseline): helped 78%, no clear effect 4%, hurt 17%, mixed 2%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add sparse global matching with 1/8-point flow compensation | X-Test-L · PSNR | 30.39 -> 30.83; 29.25 -> 29.73 | Helped | Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · 30 citations) |
| Add the quasi-binary mask instead of a binary mask | DAVIS-2017 480P · VFIPS | 81.25 -> 83.30 | Helped | Perception-Oriented Video Frame Interpolation via Asymmetric Blending (CVPR 2024 · 24 citations) |
| use a two-stage backbone rather than the linear scheme | Vimeo90K · PSNR | 32.59 -> 35.35 | Helped | IQ-VFI: Implicit Quadratic Motion Estimation for Video Frame Interpolation (CVPR 2024 · 17 citations) |
| add Reliable Pixel Sampling and Patch-Mixed Sampling to self-supervised adaptation with NR Block | Cross-domain (BS-ERGB to HS-ERGB) · PSNR | 31.53 -> 32.13 | Helped | TTA-EVF: Test-Time Adaptation for Event-based Video Frame Interpolation via Reliable Pixel and Sample Estimation (CVPR 2024 · 13 citations) |
| Use S6 instead of convolution for inter-frame modeling | X-TEST 2K · PSNR | 31.58 -> 32.84 | Helped | VFIMamba: Video Frame Interpolation with State Space Models (NeurIPS 2024 · 11 citations) |
| replace INR with U-Net | VTF and VSF · PSNR | 37.56 -> 36.96; 30.45 -> 29.96 | Hurt | Generalizable Implicit Motion Modeling for Video Frame Interpolation (NeurIPS 2024 · 10 citations) |
| replace alternative scans with window-based Hilbert scanning | Xiph-4K · PSNR | 33.13 -> 34.26; 33.80 -> 34.26; 33.99 -> 34.26; 34.01 -> 34.26; 33.30 -> 34.26 | Helped | LC-Mamba: Local and Continuous Mamba with Shifted Windows for Frame Interpolation (CVPR 2025 · 8 citations) |
| Add event guidance and multi-level reference fusion to the PCD alignment module | HS-ERGB far · PSNR | 29.92 -> 33.56 | Helped | Video Frame Interpolation via Direct Synthesis with the Event-based Reference (CVPR 2024 · 7 citations) |
| Use alternating training with the synthesis network fixed when fine-tuning the flow teacher | SNU-FILM-hard · LPIPS | 0.0445 -> 0.0419; 0.0475 -> 0.0419 | Helped | Disentangled Motion Modeling for Video Frame Interpolation (AAAI 2025 · 6 citations) |
| add the motion refinement module to TPS-only coarse motion estimation | MixiamoLine240 validation, N=5 · CD (×10^5) | 11.800 -> 9.235 | Helped | Thin-Plate Spline-based Interpolation for Animation Line Inbetweening (AAAI 2025 · 5 citations) |
| Use BiM instead of time indexing T | Vimeo90K-septuplet · LPIPS | 0.098 -> 0.070 | Helped | BiM-VFI: Bidirectional Motion Field-Guided Frame Interpolation for Video with Non-uniform Motions (CVPR 2025 · 5 citations) |
| Add Two-side Fusion of forward and backward denoised latents | EVFI · frame interpolation consistency | Helped | Repurposing Pre-trained Video Diffusion Models for Event-based Video Interpolation (CVPR 2025 · 5 citations) |
Baseline numbers that disagree
The same baseline on the same benchmark and metric, reported more than 5% apart by different papers here. Training budgets, data splits and reimplementations differ, so state which number you compare with.
| Baseline | Benchmark | Metric | Reported | Papers |
|---|---|---|---|---|
| SepConv | UCF101 | PSNR | 33.02 – 34.73 | 2 papers, e.g. CVPR 2018 and CVPR 2020 |
What reviewers will expect
Datasets and baselines named in the 8 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- Vimeo90k · 2 of 8 papers
- SNU-FILM · 2 of 8 papers
Baselines
- RIFE · 5 of 8 papers
- EMA-VFI · 5 of 8 papers
- ABME · 3 of 8 papers
- IFRNet · 3 of 8 papers
- PerVFI · 3 of 8 papers
- FILM · 2 of 8 papers
- UPRNet · 2 of 8 papers
- CAIN · 2 of 8 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 33 papers; from OpenAlex and the papers' affiliations).
- Harbin Institute of Technology · 4 papers
- Korea Advanced Institute of Science and Technology · 3 papers
- Nanjing University · 3 papers
- Northwestern Polytechnical University · 2 papers
- Wuhan University · 2 papers
- The University of Tokyo · 2 papers
- Shanghai AI Lab · 2 papers
- Shanghai Jiao Tong University · 2 papers
- SenseTime · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Learned optical flow estimation (73 papers)
- Event-based video reconstruction (63 papers)
- Spike and neuromorphic deblurring (57 papers)
Elsewhere in the topic
- Self-supervised video representation learning (167 papers)
- Video restoration with temporal priors (159 papers)
- Layered video decomposition and dynamics (140 papers)
- Stochastic video prediction (125 papers)
- Physics-inspired motion prediction (113 papers)
- Sign language and physiological video analysis (99 papers)
- Weakly supervised video anomaly detection (99 papers)
- Spatio-temporal predictive modeling (91 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).