All silhouettes are assumed to be standing upright.Points of interest (POIs) are defined as pixels or sets of pixels in a skeleton located at the endpoint of a segment. A POI may also be the intersection between two or more segments.
Purpose of POIs
Method Overview
Purpose of tracking POIs and segmentsIdentify body parts and arm activity
Improve skeletons
Improve "blobs"
Bone PropertiesTorso
- Always present in useful frames
- Usually the largest and only near-vertical bone (except head, also near-vertical)
- All limbs, including head, join to it via an "intersection" POI
Head
- Always with one intersection and one endpoint
- Near-vertical, almost a continuation of the torso
Arms
- Rejoin torso/head and become invisible at extreme angles; near-vertical only detectable if head-segment and arm(s) disappear and the torso's top endpoint jumps toward the top of the frame
- May consist of more than one segment if elbow is detectable
- Intersects torso near "top"
- Focus of hold-up detection
Legs
- Frequently consist of 2 or 3 segments (leg & 2 for feet)
- May also have large slope; range still not as high as torso/head
- Often legs join together
- Often vestigial--much like the feet, not very distinguishable in blob
- Bottom of torso
Some common configurations:
![]()
A
![]()
B
![]()
C
- head, endpoint & intersection at top, near-vertical; arm, endpoint & intersection with torso, non-vertical; torso, 2 intersections, long, high slope; legs, intersection & endpoint, high slope
- torso and vestigial legs only; torso includes endpoint
- torso only (head can be created as part of segment)
Problems
Loops: 2 segments with the same endpoints--2 intersections
- Ends of blob: 2 body segments (such as legs) mistakenly joined at one end
- Otherwise: hole in blob that shouldn't be there
Solved by eliminating such holes before skeleton calculation.
![]()
![]()
Bone Spurs:
short or illogically located bone
- Eliminate it because it is an error caused by the irregular blob boundaries
- Absorb it into the bone to which it really belongs (can still eliminate it because it offers no additional information)
![]()
Method OverviewFirst determine the POIs and segments of the skeleton.
![]()
![]()
Find the head based on the silhouette.
Head-based identification.
- Highest unchecked pixel
- Connected components downward
- Centermost above threshhold is head
- Dimensions may offer shoulder guess
![]()
Find the end of the head using three different calculations:
Sample Image and Video:
Width-Based Gradient-Based Circle-Fit
![]()
Arms must be addressed further.
- Head: reliable starting point (relatively stable, easily found)
- Torso: top attached to head
- Legs: tops attached to torso
- Arms: highly mobile, therefore highly variable
Also, the height:width ratio is calculated, and if it is
- Two levels of certainty to keep arms during appearance and rotation.
- Broken arms adjusted to simplify skeleton and improve accuracy.
greater than 3.9, legs are possible.This helps avoid marking arms in a partial silhouette as legs and helps narrow the search for arms.
greater than 3.0, legs may be possible, based on history.
less than 3.0, legs are not possible unless history strongly contradicts.
![]()
![]()
AlarmsArm positions are analyzed based on slope. The first and lowest alarm is the horizontal arm position. This is the lowest level because when an arm first becomes distinguishable from the torso, it almost always falls into this category. It is identified by an orange bar across the top (for left arm) or bottom (for right arm) of the image.
The second alarm level is the vertical arm position. An arm will move from the first alarm level to no alarm before moving into this position, although this is not yet checked by the program. A red bar down the left or right side of the image indicates a raised arm.
Globally, the highest alarm is reached when there are multiple silhouettes and both level and raised arm alarms.
August 2002
Sequence # Frames Successful ID walaim1 68 94.12% walaim2 115 93.04% walup1 104 87.5% walup2 97 93.81% mikeaim 512 92.77% mikemulti 548 91.24% TOTAL 1444 91.96% These sequences are of a single person fully visible outdoors. The "wal" sequences were of a quick mover late on an overcast afternoon in front of a wall and moving branches. The "mike" sequences were of a slower mover who rotated after walking to the center of the picture; the setting was a bright, early afternoon on grass in front of a pale wall.
Sample Videos
![]()
![]()
![]()
![]()
![]()
| Project main page
Last updated: 23 NOV 2005
Created: 9 OCT 2001