Open-vocabulary Object Retrieval

By Sergio Guadarrama, Erik Rodner, Kate Saenko, Ning Zhang, Ryan Farrell, Jeff Donahue, Trevor Darrell. 2014

95 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Caffe: Convolutional Architecture for Fast Feature Embedding (2014, methodology): “This capability is essential for tasks such as knowledge transfer [2], object detection [3], and object retrieval [5].”
Grounding of Textual Phrases in Images by Reconstruction (2015, background): “[15] predicts a class for each proposal box and constructs a word bag with all the synonyms of the class-name based on WordNet [11].”
Natural Language Object Retrieval (2015, methodology): “Similar to [10], we evaluate with “P@1-NR” corresponding to non-random top-1 precision computed on the those informative results and “P@1” corresponding to top-1 precision on all cases including non-informative results, where random guess is used.”
Gated-Attention Architectures for Task-Oriented Language Grounding (2017, neutral): “In the context of grounding language in objects and their attributes, Guadarrama et al. (2014) present a method to ground open vocabulary to objects in the environment.”
Recent Advances in Open Set Recognition: A Survey (2018, background): “There also some conceptually similar topics have been studied in other research communities such as open-vocabulary object retrieval [163], [164], open world person re-identification [165] or searching targets [166], open vocabulary scene parsing [167].”
A Joint Network for Grasp Detection Conditioned on Natural Language Commands (2021, methodology): “and textual input can then be obtained by grounding-viageneration strategy [3], [4], [12], or designing a proximitybased similarity score function in the shared image-and-text embedding space such as inner-product [1], [12] or cosine similarity [6].”
Tell me Dave: Context-sensitive grounding of natural language to manipulation instructions (2014, background): “As mentioned before, there are several works (Guadarrama et al., 2013, 2014) which aim to solve this problem.”
Entity-Enhanced Adaptive Reconstruction Network for Weakly Supervised Referring Expression Grounding (2022, unknown): “Same with previous works, if the IoU Method RefCLEF Flickr30k LRCN [49] 8.59 - Caffe-7K [54] 10.38 - GroundeR [19] 28.94 MATN [22] 13.61 33.10 VC [23] 14.11 - VC w/o α [23] 14.50 - KAC Net [20] 15.83 38.71 ARN [30] 26.”
Deep Object-Centric Representations for Generalizable Robot Learning (2017, methodology): “Similarly, using language to bridge instance matching and category matching can provide flexible object detection [6].”
Procedure Planning in Instructional Videos (2019, background): “The key challenge of the task described in Figure 1 is the open-vocabulary [5, 7, 25] state and action spaces in real-world instructional videos.”
Semi-supervised Vocabulary-Informed Learning (2016, background): “Conceptually similar, but different in formulation and task, open-vocabulary object retrieval [20] focused on retrieving objects using natural language open-vocabulary queries.”