Open-world text-specified object counting
Our objective is open-world object counting in images, where the target object class is specified by a text description. To this end, we propose CounTX, a class-agnostic, single-stage model using a transformer decoder counting head on top of pre-trained joint text-image representations. CounTX is ab...
Main Authors: | Amini-Naieni, N, Amini-Naieni, K, Han, T, Zisserman, A |
---|---|
Format: | Conference item |
Language: | English |
Published: |
British Machine Vision Association
2023
|
Similar Items
-
Instant uncertainty calibration of NERFs using a meta-calibrator
by: Amini-Naieni, N, et al.
Published: (2024) -
Policy Networks, Concepts and Approaches
in Policy Making
by: Ali khaje Naieni
Published: (2016-03-01) -
Learning To Count Objects in Images
by: Lempitsky, V, et al.
Published: (2010) -
Development and Utilization of Bridge Data of the United States for Predicting Deck Condition Rating Using Random Forest, XGBoost, and Artificial Neural Network
by: Fariba Fard, et al.
Published: (2024-01-01) -
Simultaneous Selection of the Technology and its Transfer Method from the Preferred Supplier Applying BWM and Grey ANP Methods
by: Ali Bonyadi Naieni, et al.
Published: (2020-03-01)