Research map / Video / Video segmentation and tracking
Neural video representations
Encodes videos as implicit networks with hybrid or frame-embedding designs for compression.
21 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2026) · an approach in Video segmentation and tracking, keyword cluster “compression · coding · rate”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
15 papers in 2023–24 and 3 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 3.4% to 0.6%.
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
- 2019 · BehaveNet: nonlinear embedding and Bayesian neural decoding of behavioral videos (NeurIPS 2019 · 73 citations)
- 2021 · NeRV: Neural Representations for Videos (NeurIPS 2021 · 3 citations)
- 2022 · Scalable Neural Video Representations with Learnable Positional Features (NeurIPS 2022 · 12 citations)
- 2023 · HNeRV: A Hybrid Neural Representation for Videos (CVPR 2023 · 118 citations)
- 2024 · Boosting Neural Representations for Videos with a Conditional Decoder (CVPR 2024 · 32 citations)
- 2025 · PNVC: Towards Practical INR-based Video Compression (AAAI 2025 · 14 citations)
- 2026 · CoANeRV: Coordinate-Aware Token-Space Neural Video Representation (NeurIPS 2026 · 0 citations)
Earliest papers here
- BehaveNet: nonlinear embedding and Bayesian neural decoding of behavioral videos (NeurIPS 2019 · 73 citations)
- NeRV: Neural Representations for Videos (NeurIPS 2021 · 3 citations)
- Scalable Neural Video Representations with Learnable Positional Features (NeurIPS 2022 · 12 citations)
What has been tried
19 papers here report 223 measured changes (a component added, removed or swapped, compared with a baseline): helped 73%, no clear effect 2%, hurt 20%, mixed 4%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add TAT residual blocks | Bunny (720p), 300 epochs · PSNR | 39.12 -> 41.09 | Helped | Boosting Neural Representations for Videos with a Conditional Decoder (CVPR 2024 · 32 citations) |
| Add reparameterization during pretraining | UVG and MCL-JCV · BD-rate | 6.45% (UVG); 6.34% (MCL-JCV) coding loss when removed | Helped | PNVC: Towards Practical INR-based Video Compression (AAAI 2025 · 14 citations) |
| add two-pass progressive scaling | sport1 (NHR) · PSNR | Helped | TeTriRF: Temporal Tri-Plane Radiance Fields for Efficient Free-Viewpoint Video (CVPR 2024 · 14 citations) | |
| Replace dual-axis layer Gaussian model with per-tensor conditional Gaussian model | UVG · BD-rate PSNR | 0.00% -> 13.04% | Hurt | NVRC: Neural Video Representation Compression (NeurIPS 2024 · 11 citations) |
| add temporal guidance (progressive frame learning) | HMDB-51 · PSNR (Step1/Step3) | 16.89/19.43 -> 17.3/19.31 | Mixed | MetaNeRV: Meta Neural Representations for Videos with Spatial-Temporal Guidance (AAAI 2025 · 5 citations) |
| use frame-specific embeddings only instead of combining embeddings and GOP-specific feature grids | UVG, 720p, first 300 frames, 3M parameters (Ready.) · PSNR | 32.23 -> 32.45 | Helped | Combining Frame and GOP Embeddings for Neural Video Representation (CVPR 2024 · 5 citations) |
| add implicit motion representation to explicit representation and alignment | Flower · L1 | 0.0642 -> 0.0544 | Helped | Implicit Motion Function (CVPR 2024 · 4 citations) |
What reviewers will expect
Datasets and baselines named in the 10 papers of this approach from 2024–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- UVG · 4 of 10 papers
- MCL_JCV · 2 of 10 papers
- Bunny · 2 of 10 papers
Baselines
- NeRV · 4 of 10 papers
- FFNeRV · 4 of 10 papers
- E-NeRV · 3 of 10 papers
- HNeRV · 3 of 10 papers
- H.264 · 3 of 10 papers
- DCVC-DC · 3 of 10 papers
- DCVC-HEM · 2 of 10 papers
- C3 · 2 of 10 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 18 papers; from OpenAlex and the papers' affiliations).
- Zhejiang University · 3 papers
- University of Maryland, College Park · 2 papers
- Meta · 2 papers
- Studios · 2 papers
- Disney Research · 2 papers
- ETH Zurich · 2 papers
- University of Bristol · 2 papers
Use it
Download BibTeX of 20 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 video compression (68 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).