Exploring Ordinal Bias in Action Recognition for Instructional Videos
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915655109836800 |
|---|---|
| author | Kim, Joochan Jung, Minjoon Zhang, Byoung-Tak |
| author_facet | Kim, Joochan Jung, Minjoon Zhang, Byoung-Tak |
| contents | Action recognition models have achieved promising results in understanding instructional videos. However, they often rely on dominant, dataset-specific action sequences rather than true video comprehension, a problem that we define as ordinal bias. To address this issue, we propose two effective video manipulation methods: Action Masking, which masks frames of frequently co-occurring actions, and Sequence Shuffling, which randomizes the order of action segments. Through comprehensive experiments, we demonstrate that current models exhibit significant performance drops when confronted with nonstandard action sequences, underscoring their vulnerability to ordinal bias. Our findings emphasize the importance of rethinking evaluation strategies and developing models capable of generalizing beyond fixed action patterns in diverse instructional videos. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_06580 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Exploring Ordinal Bias in Action Recognition for Instructional Videos Kim, Joochan Jung, Minjoon Zhang, Byoung-Tak Computer Vision and Pattern Recognition Artificial Intelligence Action recognition models have achieved promising results in understanding instructional videos. However, they often rely on dominant, dataset-specific action sequences rather than true video comprehension, a problem that we define as ordinal bias. To address this issue, we propose two effective video manipulation methods: Action Masking, which masks frames of frequently co-occurring actions, and Sequence Shuffling, which randomizes the order of action segments. Through comprehensive experiments, we demonstrate that current models exhibit significant performance drops when confronted with nonstandard action sequences, underscoring their vulnerability to ordinal bias. Our findings emphasize the importance of rethinking evaluation strategies and developing models capable of generalizing beyond fixed action patterns in diverse instructional videos. |
| title | Exploring Ordinal Bias in Action Recognition for Instructional Videos |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2504.06580 |