Skip to content

Repository files navigation

Awesome Video Summarization

A research handbook and practical learning guide to video summarization: classical and supervised methods, weak/semi/self-supervision, unsupervised and reinforcement learning, training-free selection, multimodal foundation models, and query or user-conditioned systems. It covers keyframes, keyshots, textual and multimodal summaries, with source-linked code, datasets and protocol-aware benchmarks.

Last literature audit: 2026-09-19. Coverage is selective and dated; the ledger and latest audit distinguish newly verified entries from inherited technical audits. Source availability does not imply successful reproduction.

Start learning: Learning path · Find a paper: Catalog · Find code: Implementations · Compare results: Benchmarks

Read the handbook

Foundation Method families Practice and research
History and field overview Classical and supervised Task-specific settings
Taxonomy and formulations Reconstruction and generative Protocol-separated benchmarks
Mathematical foundations RL and heuristic selection Code, features and weights
Evaluation and failure modes Contrastive and self-supervised Open problems
Datasets and annotations Graphs, attention and Transformers Coverage and research gaps
Hands-on learning path Multimodal, VLM and Video-LLM Changelog

Choose a task, then a protocol

Axis Examples
Supervision Supervised, weak/semi/self-supervised, unsupervised, training-free, zero/few-shot
Output Storyboard, keyshot skim, highlight, text, multimodal, timestamp/event summary
Setting Generic, query-focused, personalized, multi-video, egocentric, long-form, online
Mechanism Clustering, sparse/submodular/DPP, recurrent, reconstruction/GAN, RL, graph/attention, contrastive, VLM/LLM, diffusion

These axes are independent. The four original unsupervised families remain a sub-taxonomy. Supervised methods are core reading, including methods whose frozen encoders or auxiliary contrastive losses can obscure their human-label dependence.

Datasets and results

SumMe and TVSum support classic importance-to-skim studies; VideoXum supports visual and textual summaries; query and egocentric resources have their own annotations and metrics. MoSu adds behavior-derived multimodal importance targets, TRINITY separates Event/Emotion/Nature saliency, KnowVis generates pedagogical images, and Multi-VidSum pairs keyframes with captions. QA and grounding resources remain explicitly adjacent. See dataset cards for licensing, availability, features and unresolved fields.

A score is meaningful only with its split, training data, features, sampling, segmentation, shot values, budget solver, reference aggregation and metric. The generated benchmark catalog isolates unknown/incompatible protocols and distinguishes author-reported values from reproduced ones. It does not assert a field-wide winner.

Recent starting points include TRINITY, SGWIB, metadata-conditioned zero-shot highlight detection, KnowVis, agentic narrative summaries, multimodal sports systems, TripleSumm and CoE. The modern-method chapter explains their actual training signals and output differences; the latest audit preserves count conflicts, restricted sources and withheld numerical claims.

Run a first experiment

No model or dataset download is needed for the synthetic decoder example:

python3 examples/summary_baselines.py --demo

Change shot lengths, scores and budget to see how knapsack and user-reference aggregation affect the output. Continue with the learning path and source-audited resources, including official implementations, author presentations, tutorials, feature archives and checkpoints.

Contribute and maintain

CONTRIBUTING.md defines evidence, schemas and validation. Edit registries under data/, then run python3 scripts/generate_catalog.py. The documentation workflow checks schemas, generated files, Markdown, mathematics, Mermaid and tests. The weekly external-link workflow reports restricted access separately from missing links.

Citation and license

Use CITATION.cff to cite this handbook with the revision you used, and cite original papers/datasets for their methods and results. The repository is MIT licensed; linked code, datasets, media and model weights retain their own licenses.

About

A curated, mathematically rigorous survey and resource catalog for unsupervised, self-supervised, and multimodal foundation-model video summarization.

Topics

Resources

Contributing

Stars

4 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages