What is he best known for?
The fields of study he is best known for:
- Artificial intelligence
- Algorithm
- Machine learning
Hervé Jégou spends much of his time researching Artificial intelligence, Pattern recognition, Image retrieval, Image and Nearest neighbor search.
Many of his studies on Artificial intelligence involve topics that are commonly interrelated, such as Computer vision.
His studies deal with areas such as Hamming embedding, Visual Word, Quantization and Rank as well as Pattern recognition.
Hervé Jégou has included themes like Inverted index and Geometric transformation in his Visual Word study.
His Image study incorporates themes from Measure, Representation and Data mining.
The concepts of his Nearest neighbor search study are interwoven with issues in Short Code, Quantization and Search engine indexing.
His most cited work include:
- Product Quantization for Nearest Neighbor Search (1692 citations)
- Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search (1663 citations)
- Improving Bag-of-Features for Large Scale Image Search (679 citations)
What are the main themes of his work throughout his whole career to date?
His primary areas of study are Artificial intelligence, Pattern recognition, Image, Search engine indexing and Algorithm.
His work carried out in the field of Artificial intelligence brings together such families of science as Machine learning and Computer vision.
His Pattern recognition research is multidisciplinary, incorporating elements of Scale, Fisher vector, Kernel and Visual Word.
As a part of the same scientific family, Hervé Jégou mostly works in the field of Visual Word, focusing on Inverted index and, on occasion, Geometric transformation and Binary number.
The various areas that Hervé Jégou examines in his Image study include Object, Data mining and Convolutional neural network.
His work on Best bin first as part of general Nearest neighbor search study is frequently connected to Locality-sensitive hashing, therefore bridging the gap between diverse disciplines of science and establishing a new relationship between them.
He most often published in these fields:
- Artificial intelligence (60.80%)
- Pattern recognition (36.36%)
- Image (25.00%)
What were the highlights of his more recent work (between 2019-2021)?
- Artificial intelligence (60.80%)
- Transformer (5.68%)
- Contextual image classification (10.23%)
In recent papers he was focusing on the following fields of study:
Hervé Jégou mainly investigates Artificial intelligence, Transformer, Contextual image classification, Image and Algorithm.
His Artificial intelligence research includes elements of Machine learning and Metric.
His Transformer research also works with subjects such as
- Convolutional neural network that connect with fields like Scale, State, FLOPS, Parallel processing and Hybrid neural network,
- Computer engineering that intertwine with fields like Range.
His Image study incorporates themes from Training set and Pattern recognition.
In his works, Hervé Jégou conducts interdisciplinary research on Pattern recognition and Training.
His Quantization and Gradient based algorithm study in the realm of Algorithm connects with subjects such as Sampling.
Between 2019 and 2021, his most popular works were:
- Training data-efficient image transformers & distillation through attention (95 citations)
- Fixing the train-test resolution discrepancy: FixEfficientNet (46 citations)
- And the Bit Goes Down: Revisiting the Quantization of Neural Networks (34 citations)
In his most recent research, the most cited papers focused on:
- Artificial intelligence
- Machine learning
- Algorithm
Hervé Jégou mostly deals with Contextual image classification, Algorithm, Quantization, Image and Artificial intelligence.
His research integrates issues of Machine learning, Model compression, Product quantization and Transformer in his study of Contextual image classification.
His research in the fields of Reconstruction error and Vector quantization overlaps with other disciplines such as Memory footprint and Network architecture.
In most of his Quantization studies, his work intersects topics such as Inference.
His Image study combines topics in areas such as Training set, Pattern recognition, Overfitting and Resolution.
His work often combines Artificial intelligence and Distillation studies.
This overview was generated by a machine learning system which analysed the scientist’s
body of work. If you have any feedback, you can
contact us
here.
