Learning to count anything: reference-less class-agnostic counting with weak supervision
Current class-agnostic counting methods can generalise to unseen classes but usually require reference images to define the type of object to be counted, as well as instance annotations during training. Reference-less class-agnostic counting is an emerging field that identifies counting as, at its c...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
2023
|
| _version_ | 1824458584756322304 |
|---|---|
| author | Hobley, M Prisacariu, V |
| author_facet | Hobley, M Prisacariu, V |
| author_sort | Hobley, M |
| collection | OXFORD |
| description | Current class-agnostic counting methods can
generalise to unseen classes but usually require reference
images to define the type of object to be counted, as well
as instance annotations during training. Reference-less
class-agnostic counting is an emerging field that identifies
counting as, at its core, a repetition-recognition task. Such
methods facilitate counting on a changing set composition.
We show that a general feature space with global context
can enumerate instances in an image without a prior on
the object type present. Specifically, we demonstrate that
regression from vision transformer features without pointlevel supervision or reference images is superior to other
reference-less methods and is competitive with methods
that use reference images. We show this on the current
standard few-shot counting dataset FSC-147. We also
propose an improved dataset, FSC-133, which removes
errors, ambiguities, and repeated images from FSC-147
and demonstrate similar performance on it. To the best
of our knowledge, we are the first weakly-supervised
reference-less class-agnostic counting method. |
| first_indexed | 2024-03-07T08:17:03Z |
| format | Conference item |
| id | oxford-uuid:52c65dce-08e0-4566-abdb-66f4d45d66ef |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:28:13Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | oxford-uuid:52c65dce-08e0-4566-abdb-66f4d45d66ef2024-12-09T16:04:30ZLearning to count anything: reference-less class-agnostic counting with weak supervisionConference itemhttp://purl.org/coar/resource_type/c_c94fuuid:52c65dce-08e0-4566-abdb-66f4d45d66efEnglishSymplectic Elements2023Hobley, MPrisacariu, VCurrent class-agnostic counting methods can generalise to unseen classes but usually require reference images to define the type of object to be counted, as well as instance annotations during training. Reference-less class-agnostic counting is an emerging field that identifies counting as, at its core, a repetition-recognition task. Such methods facilitate counting on a changing set composition. We show that a general feature space with global context can enumerate instances in an image without a prior on the object type present. Specifically, we demonstrate that regression from vision transformer features without pointlevel supervision or reference images is superior to other reference-less methods and is competitive with methods that use reference images. We show this on the current standard few-shot counting dataset FSC-147. We also propose an improved dataset, FSC-133, which removes errors, ambiguities, and repeated images from FSC-147 and demonstrate similar performance on it. To the best of our knowledge, we are the first weakly-supervised reference-less class-agnostic counting method. |
| spellingShingle | Hobley, M Prisacariu, V Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title | Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title_full | Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title_fullStr | Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title_full_unstemmed | Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title_short | Learning to count anything: reference-less class-agnostic counting with weak supervision |
| title_sort | learning to count anything reference less class agnostic counting with weak supervision |
| work_keys_str_mv | AT hobleym learningtocountanythingreferencelessclassagnosticcountingwithweaksupervision AT prisacariuv learningtocountanythingreferencelessclassagnosticcountingwithweaksupervision |