Science Explorer Back to mapMap
Discussion
Topic · Computer Vision and Pattern Recognition

Human Pose and Action Recognition

Human Pose and Action Recognition is a research topic within Computer Vision and Pattern Recognition. Science Explorer counts 29k research works in it since 1969. 9.5% of them reached the world's top 10% most cited for their field and year.

This cluster of papers focuses on the development and application of deep learning techniques for human action recognition and pose estimation. It covers topics such as spatiotemporal feature learning, convolutional networks, 3D human pose estimation, skeleton-based recognition, and video classification. The research aims to advance the understanding and accurate detection of human actions in various environments.

  • Action Recognition
  • Pose Estimation
  • Spatiotemporal Features
  • Convolutional Networks
  • 3D Human Pose
  • Skeleton-Based Recognition
  • Video Classification
  • Deep Learning
  • Human Activity Analysis
  • Temporal Convolutional Networks
Research works
29k
fractional, since 1969
In the world top 10%
2.8k
per year above
Top-10% rate
9.5%
share of its works in the world top 10%
Growth, 2018-21 → 2022-25
+53%
the tick is no change
Largest producer
China
56% of the 10 largest, 2022-now
Human Pose and Action Recognition

Which countries lead Human Pose and Action Recognition research?

By volume, China and India publish the most (5.4k and 1.3k works in 2022-now).

By volume, 2022-now

  1. 1 China 5.4k works
  2. 2 India 1.3k works
  3. 3 United States 1k works
  4. 4 Japan 437 works
  5. 5 South Korea 420 works
  6. 6 United Kingdom 305 works
  7. 7 Germany 295 works
  8. 8 Italy 188 works
  9. 9 Canada 174 works
  10. 10 France 168 works

How concentrated that is

The same countries as shares of everything the list above accounts for. A node where two countries do two thirds of the work and one spread evenly across twelve read alike as a ranking and not at all alike here.

China: 55.7%India: 13.3%United States: 10.4%Japan: 4.5%6 others listed: 16.0%56%largest
China5,396 · 55.7%India1,291 · 13.3%United States1,008 · 10.4%Japan437 · 4.5%6 others listed1,551 · 16.0%

Shares of the rows listed above, not of the whole node.

Which institutions lead Human Pose and Action Recognition research?

By volume in 2022-now, Chinese Academy of Sciences publishes the most Human Pose and Action Recognition research, followed by Tsinghua University and Shanghai Jiao Tong University.

By volume, 2022-now

  1. 1 Chinese Academy of SciencesChina 201 works
  2. 2 Tsinghua UniversityChina 108 works
  3. 3 Shanghai Jiao Tong UniversityChina 92 works
  4. 4 Beijing University of Posts and TelecommunicationsChina 91 works
  5. 5 Zhejiang UniversityChina 86 works
  6. 6 University of Science and Technology of ChinaChina 82 works
  7. 7 Peking UniversityChina 80 works
  8. 8 Sun Yat-sen UniversityChina 71 works
  9. 9 Beijing Institute of TechnologyChina 69 works
  10. 10 University of Electronic Science and Technology of ChinaChina 67 works

Who are the leading researchers in Human Pose and Action Recognition?

The most-cited researchers publishing on Human Pose and Action Recognition include Kaiming He, Yoshua Bengio and Geoffrey E. Hinton.

  1. 1 Kaiming He Israel 562k citations
  2. 2 Yoshua Bengio Canada 476k citations
  3. 3 Geoffrey E. Hinton Canada 454k citations
  4. 4 Ross Girshick Israel 356k citations
  5. 5 Xiangyu Zhang 327k citations
  6. 6 Jian Sun 281k citations
  7. 7 Xiaogang Wang China 260k citations
  8. 8 Yann LeCun United States 256k citations
  9. 9 Andrew Zisserman United Kingdom 255k citations
  10. 10 Li Fei-Fei United States 225k citations

Ranked by citations received across their whole record, among researchers with at least three works on this topic.

What are the most cited papers in Human Pose and Action Recognition?

The most cited paper in Human Pose and Action Recognition is "Densely Connected Convolutional Networks" (Gao Ming Huang, Zhuang Liu, Laurens van der Maaten et al., 2017), with 45,976 citations. Next come "Histograms of Oriented Gradients for Human Detection" (32,222) and "Mask R-CNN" (29,752).

Most cited papers
#PaperReleasedCitations
1 Densely Connected Convolutional Networks Gao Ming Huang, Zhuang Liu, Laurens van der Maaten et al. Conference paper Jul 2017 45,976
2 Histograms of Oriented Gradients for Human Detection Navneet Dalal, Bill Triggs Conference paperOpen access 27 Jul 2005 32,222
3 Mask R-CNN Kaiming He, Georgia Gkioxari, Piotr Dollár et al. Conference paperOpen access Oct 2017 29,752
4 Non-local Neural Networks Xiaolong Wang, Ross Girshick, Abhinav Gupta et al. Conference paper Jun 2018 11,465
5 Learning Spatiotemporal Features with 3D Convolutional Networks Du Tran, Lubomir Bourdev, Rob Fergus et al. Conference paper Dec 2015 9,921
All 98 most cited papers, with filters and sorting

Where is Human Pose and Action Recognition research done?

The largest centres of Human Pose and Action Recognition research in 2022-now are Beijing (China), Shanghai (China), Xi'an (China) and Nanjing (China). Among places with at least 20 works in it, it is an unusually large share of all research in Guildford.

Largest cities, 2022-now

  1. 1 BeijingChina 1.1k works
  2. 2 ShanghaiChina 382 works
  3. 3 Xi'anChina 291 works
  4. 4 NanjingChina 286 works
  5. 5 GuangzhouChina 231 works
  6. 6 HangzhouChina 226 works
  7. 7 WuhanChina 225 works
  8. 8 SeoulSouth Korea 199 works
  9. 9 ShenzhenChina 194 works
  10. 10 TokyoJapan 169 works

Where it is the local speciality

  1. GuildfordGB · 20.9 works6.3×
← less than its size predicts1×more →

Location quotient: how much more of its research is in Human Pose and Action Recognition than the world average.

See Human Pose and Action Recognition on the map

Where is the best place to study Human Pose and Action Recognition?

Among universities, judged by research, Air University, Nanyang Technological University and Peking University score highest, combining excellence, specialisation, size, growth and international reach. Research strength is one signal when choosing where to study; it does not measure teaching.

0%20%40%mean 21.82%fractional works in this field (log) →share of work in the world top 10% →Air UniversityPeking UniversityNanyang Technologica…Tsinghua University
above the meannear itbelow it

One dot per university in the first ten of the table below, at the default weights. The upper left is the interesting corner: small places doing unusually strong work.

Excellenceshare of its work in the world top 10%, 2022-now and all time averaged
40%
Specialisationhow much of its work is in this field, counted up to 10× the world share
10%
Sizeits output in this field, on a log scale
30%
Growthoutput growth, counted from -20% to +20%
10%
Internationalshare of its work with a co-author abroad
10%
Each slider sets a weight; the share it gets of the score is beside it.
#UniversityScoreTop 10%SpecialisationWorksGrowth
1Air UniversityPakistan 75.028.0%29.8×18 +158.3%
2Nanyang Technological UniversitySingapore 72.821.9%6.4×52 +47.0%
3Peking UniversityChina 72.422.1%3.8×80 +46.9%
4Tsinghua UniversityChina 70.317.3%4.6×108 +35.5%
5University of Hong KongChina 70.224.5%3.5×33 +178.0%
6National University of SingaporeSingapore 68.118.6%3.8×46 +61.7%
7Stanford UniversityUnited States 67.823.6%1.8×27 +19.1%
8Chinese University of Hong KongChina 66.831.1%2.9×22 -13.7%
9University of Science and Technology of ChinaChina 65.913.5%6.8×82 +27.4%
10Zhejiang UniversityChina 65.517.6%3.1×86 +58.9%

Universities only. The score blends excellence (40%), size (30%), specialisation (10%), growth (10%), international reach (10%), 2022-now; growth compares 2018-21 with 2022-25 and counts up to +20%, since a jump from a small base is mostly noise.

Every institution in Human Pose and Action Recognition, country by country

Is Human Pose and Action Recognition research growing?

Output in 2022-2025 was 53% higher than in 2018-2021, peaking in 2025. The fastest-growing topics are Human Pose and Action Recognition.

still counting19801990200020102020
grewheldshrank

The same series as a ribbon: one cell per year, darker for more. The line above answers how much; this answers when.

Its share of all research

Growing output can still be a shrinking share: all of science grows. This is Human Pose and Action Recognition as a percentage of every work published that year, 0.047% in 2024.

still counting0.020%0.040%0.060%Human Pose and Action Recognition0.050% in 202519801990200020102020

Fractional works of Human Pose and Action Recognition over all fractional works that year. How the balance of science has shifted.

How much of it reaches the world top 10%

Year by year.

still counting5%10%15%Human Pose and Action Recognition9.6% in 2025200020102020

Of each year's works in it, the share that reached the world's 10% most cited for their field and year. A field is the world for its own topics, so there is no other line to hold it against.

Which topics inside it are moving

Growth and decline on one axis around a shared zero. Two lists side by side hide the thing that matters: whether the growth dwarfs the decline, or the other way round.

Discussion

Ask a question or share a note about Human Pose and Action Recognition. Posts are plain text and can be reported.