Finding Tiny Faces
(CVPR 2017)
Peiyun Hu, Deva Ramanan
Robotics Institute
Carnegie Mellon University
1 /18
2 /18
outline
• 1.Introduction
• 2.exploring context and resolution
• 3.approach:scale-specific detection
• 4.experiment
• 5.conclusion
3 /18
1.Introduction
• Though tremendous strides have been made in object recognition, one of the
remaining open challenges is detecting small objects. We explore three aspects of
the problem in the context of face detection: the role of scale invariance,image
resolution and contextual reasoning.
• Scale invariance is a fundamental property of almost all current recognition and
object detection systems. But from a practical perspective, scale-invariance cannot
hold for sensors with finite resolution: the cues for recognizing a 300px tall face
are undeniably different that those for recognizing a 3px tall face.
4 /18
• what should the size of the template be?
• we want a small template that can detect small faces and a large template that can
exploit detailed features to increase accuracy. So,we train separate detectors
tuned for different scales .
• May suffer from lack of training data for individual scales and inefficiency from
running a large number of detectors at test time.
• To address,we train and run scale-specific detectors in a multitask fashion : they
make use of features defined over multiple layers of single (deep) feature hierarchy.
• While such a strategy results in detectors of high accuracy for large objects,finding
small things is still challenging.
5 /18
• How to generalize pre-trained networks?
6 /18
• How best to encode context?
Finding small objects is challenging because there is little signal to exploit. Hence we argue that one
must use image evidence beyond the object extent. This is often formulated as “context”.
7 /18
2.exploring context and resolution
2.1. Context
Figure 4: The green box
represents the actual face
size, while dotted boxes
represent receptive fields
associated with features
from different layers (cyan =
res2, light-blue = res3,dark-
blue = res4, black = res5).
Same colors are used in
Figures 5 and 7.
8 /18