AuroraCap
Video Detailed Captioning is a key task which aims to generate compre- hensive and coherent textual descriptions of video content, benefiting both video understanding and generation.
We propose AuroraCap, a simple video caption baseline based on a large vision-language model. We follow the simplest architecture, similar to LLaVA, without additional parameters for temporal modeling. To address the overhead caused by lengthy video sequences, we implemented a token merging strategy, reducing the num- ber of visual tokens input to the LLM to just 1% of the original amount. Surprisingly, we found that this strategy results in almost no performance drop. AuroraCap shows advancing performance on various video and image captioning benchmarks compared to existing models.
You can use AuroraCap to download our model.
1 maintainer
Metadata
1 maintainer
Release files for auroracap 0.0.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| auroracap-0.0.5.tar.gz | 45.4 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| auroracap-0.0.5-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 107.7 kB
Release files / auroracap-0.0.5.tar.gz
| Download URL | auroracap-0.0.5.tar.gz |
|---|---|
| Size | 45.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
18f8abf4eb3c5c826cf3fa7a5c4073c304a4b139b7f348ab03ed8b2981056795
|
|
BLAKE2b-256 checksum How to use checksums |
4e5b1f4025ea81aacbaea4afbca42b78cdb5c62edc36b80eefef8c76b8b090df
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.12.5
|
Release files / auroracap-0.0.5-py3-none-any.whl
| Download URL | auroracap-0.0.5-py3-none-any.whl |
|---|---|
| Size | 62.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
053acda0feefecb465b71ff1ba7a42921b6d7f2aba117a7a21f529bb63990ac0
|
|
BLAKE2b-256 checksum How to use checksums |
dc2954feb591e361cdab0651f1fe72952b4f4d7569e7f6815161dfccfd6238e0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.12.5
|
Security for auroracap 0.0.5
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
