Computer Vision Academic Program

The Sighthound Academic Program makes state-of-the-art computer vision and machine learning capabilities available to students for research purposes.

Program Features

Take advantage of detection and recognition suites:

Detection Suite
  • Face Detection
  • People Detection
  • Vehicle Detection
  • License Plate Detection
Face Suite
  • Face Authentication/Recognition
  • Facial Landmarks
  • Facial Attributes
Vehicle Suite
  • Vehicle Make/Model Recognition
  • Vehicle Color Recognition
  • License Plate Recognition

Program Perks

Easy to use, highly accurate, and FREE. For students only.

250,000 Detection and Recognition API Calls

Students will receive 250,000 free API calls per month (Regular cost: $2,388 per year).

Pre-Computed Benchmark Features

Save time with benchmark extractions done for you.


Students receive full documentation and sample scripts for ease of use.

Cash Awards

Cash prizes for the best paper and most citations each year at CVPR.

Why use Sighthound?

Because it’s more accurate and easier to use than anything else out there. And for students, it’s free. See how Sighthound compares to some sample benchmarks here:

Real Age Estimation

Methods MAE
Sighthound 5.76
Rothe et al. [1] 7.34
Microsoft [2] 7.62
Kairos 10.57
Face++ [3] 11.04

Gender Recognition

Methods Accuracy
Sighthound 91.00%
Microsoft [2] 90.86%
Rothe et al. [1] 88.75%
Levi and Hassner [4] 86.80%
Kairos [5] 84.66%
Face++ [3] 83.04%

Car Classification

Methods Accuracy (top-1)
Sighthound 93.6%
Krause et al. [6] 92.8%
Lin et al. [7] 91.3%
Zhang et al. [8] 88.4%
Xie et al. [9] 86.3%
Gosselin et al. [10] 82.7%

Extra Features

Sighthound gives you powerful tools not normally available to commercial customers by exposing the features used for computation. Eligible participants are given access to scripts in Python and Matlab that use Sighthound’s deep detection, classification and recognition tools. We have also extracted features for the most popular data sets needed by students so you can fast-track your research.

Examples of datasets used include:
Age Classification:
ChaLearn; Group Dataset; Adience
Face Recognition:
Labeled Faces in the Wild (LFW); Mega Face
Car Recognition:
Stanford Car Dataset; CompCar Dataset
Object Detection:
MOT Benchmark; DukeMTMC Dataset; Action recognition dataset (UCF101 and etc)

In addition to readily available feature sets, we provide students with a free Sighthound Cloud account which grants you the ability to run 3 million free API calls per year. Our APIs are easy to build and we provide documentation for those who aren't familiar with building APIs. For example, to return face and person objects and atrributes, your API might look like this:

    # Create Sighthound Cloud Structure
    sighthound = SighthoundCloudApi()

    # Call detection API and log result
    sighthound.detectObjectsAndAttributes(image, ['face', 'person'])
Visit the Technology page to find links to Sighthound’s published work on arXiv.

Interested in Joining?

Complete the form below to register for the Sighthound Academic Program. You must be a full-time student with an e-mail address from a recognized academic institution. The Sighthound Academic Program may not be used for any commercial purpose. Approval normally takes up to 72 hours and students will get a dedicated Sighthound Cloud account with downloadable tools and features.

Academic Program Registration

1. Rothe, R., Radu Timofte, L.V.G.: Dex: Deep expectation of apparent age from asingle image. In: International Conference on Computer Vision (ICCV). (2015)
2. Microsoft-Face-API.: (
3. Face++.: (
4. Levi, G., Hassner., T.: Age and gender classification using convolutional neural networks. In: CVPRW. (2015)
5. Kairos.: (
6. Krause, Jonathan, e.a.: Fine-grained recognition without part annotations. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015)
7. Lin, T.Y., RoyChowdhury, A., Maji., S.: Bilinear cnn models for fine-grained visual recognition. In: Proceedings of the IEEE International Conference on Computer Vision. (2015)
8. et al., X.Z.: Embedding label structures for fine-grained feature representation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2016)
9. Xie, Saining, e.a.: Hyper-class augmented and regularized deep learning for finegrained image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. (2015)
10. Gosselin, P.H., Murray, N., Jegou, H., Perronnin., F.: Re-visiting the fisher vector for fine-grained classification. In: Pattern Recognition Letters. (2014)