Automated markerless pose estimation in freely moving macaques with OpenMonkeyStudio

The rhesus macaque is an important model species in several branches of science, including neuroscience, psychology, ethology, and medicine. The utility of the macaque model would be greatly enhanced by the ability to precisely measure behavior in freely moving conditions. Existing approaches do not provide sufficient tracking. Here, we describe OpenMonkeyStudio, a deep learning-based markerless motion capture system for estimating 3D pose in freely moving macaques in large unconstrained environments. Our system makes use of 62 machine vision cameras that encircle an open 2.45 m × 2.45 m × 2.75 m enclosure. The resulting multiview image streams allow for data augmentation via 3D-reconstruction of annotated images to train a robust view-invariant deep neural network. This view invariance represents an important advance over previous markerless 2D tracking approaches, and allows fully automatic pose inference on unconstrained natural motion. We show that OpenMonkeyStudio can be used to accurately recognize actions and track social interactions.

OpenMonkeyStudio: Automated Markerless Pose Estimation in Freely Moving Macaques

The rhesus macaque is an important model species in several branches of science, including neuroscience, psychology, ethology, and several fields of medicine. The utility of the macaque model would be greatly enhanced by the ability to precisely measure its behavior, specifically, its pose (position of multiple major body landmarks) in freely moving conditions. Existing approaches do not provide sufficient tracking. Here, we describe OpenMonkeyStudio, a novel deep learning-based markerless motion capture system for estimating 3D pose in freely moving macaques in large unconstrained environments. Our system makes use of 62 precisely calibrated and synchronized machine vision cameras that encircle an open 2.45m×2.45m×2.75m enclosure. The resulting multiview image streams allow for novel data augmentation via 3D reconstruction of hand-annotated images that in turn train a robust view-invariant deep neural network model. This view invariance represents an important advance over previous markerless 2D tracking approaches, and allows fully automatic pose inference on unconstrained natural motion. We show that OpenMonkeyStudio can be used to accurately recognize actions and track two monkey social interactions without human intervention. We also make the training data (195,228 images) and trained detection model publicly available.

Can Convolution Neural Network (CNN) Triumph in Ear Recognition of Uniform Illumination Invariant?

Current deep convolution neural network (CNN) has shown to achieve superior performance on a number of computer vision tasks such as image recognition, classification and object detection. The deep network was also tested for view-invariance, robustness and illumination invariance. However, the CNN architecture has thus far only been tested on non-uniform illumination invariant. Can CNN perform equally well for very underexposed or overexposed images or known as uniform illumination invariant? This is the gap that we are addressing in this paper. In our work, we collected ear images under different uniform illumination conditions with lumens or lux values ranging from 2 lux to 10,700 lux. A total of 1,100 left and right ear images from 55 subjects are captured under natural illumination conditions. As CNN requires considerably large amount of data, the ear images are further rotated at every 5o angles to generate 25,300 images. For each subject, 50 images are used as validation/testing dataset, while the remaining images are used as training datasets. Our proposed CNN model is then trained from scratch and validation and testing results showed recognition accuracy of 97%. The results showed that 100% accuracy is achieved for images with lumens ranging above 30 but having problem with lumens less than 10 lux

Effect of silhouetting and inversion on view invariance in the monkey inferotemporal cortex

We effortlessly recognize objects across changes in viewpoint, but we know relatively little about the features that underlie viewpoint invariance in the brain. Here, we set out to characterize how viewpoint invariance in monkey inferior temporal (IT) neurons is influenced by two image manipulations—silhouetting and inversion. Reducing an object into its silhouette removes internal detail, so this would reveal how much viewpoint invariance depends on the external contours. Inverting an object retains but rearranges features, so this would reveal how much viewpoint invariance depends on the arrangement and orientation of features. Our main findings are 1) view invariance is weakened by silhouetting but not by inversion; 2) view invariance was stronger in neurons that generalized across silhouetting and inversion; 3) neuronal responses to natural objects matched early with that of silhouettes and only later to that of inverted objects, indicative of coarse-to-fine processing; and 4) the impact of silhouetting and inversion depended on object structure. Taken together, our results elucidate the underlying features and dynamics of view-invariant object representations in the brain. NEW & NOTEWORTHY We easily recognize objects across changes in viewpoint, but the underlying features are unknown. Here, we show that view invariance in the monkey inferotemporal cortex is driven mainly by external object contours and is not specialized for object orientation. We also find that the responses to natural objects match with that of their silhouettes early in the response, and with inverted versions later in the response—indicative of a coarse-to-fine processing sequence in the brain.

Dynamics of 3D view invariance in monkey inferotemporal cortex

Rotations in depth are challenging for object vision because features can appear, disappear, be stretched or compressed. Yet we easily recognize objects across views. Are the underlying representations view invariant or dependent? This question has been intensely debated in human vision, but the neuronal representations remain poorly understood. Here, we show that for naturalistic objects, neurons in the monkey inferotemporal (IT) cortex undergo a dynamic transition in time, whereby they are initially sensitive to viewpoint and later encode view-invariant object identity. This transition depended on two aspects of object structure: it was strongest when objects foreshortened strongly across views and were similar to each other. View invariance in IT neurons was present even when objects were reduced to silhouettes, suggesting that it can arise through similarity between external contours of objects across views. Our results elucidate the viewpoint debate by showing that view invariance arises dynamically in IT neurons out of a representation that is initially view dependent.