3D
Model-based Object Recognition and Segmentation in Cluttered
Scenes
Abstract
Viewpoint independent recognition of free-form objects and their
segmentation in the presence of clutter and occlusions is a challenging
task. We present a novel 3D model-based algorithm which performs this
task automatically and efficiently. A 3D model of an object is
automatically constructed offline from its multiple unordered range
images (views). These views are converted into multidimensional table
representations (which we refer to as tensors). Correspondences are
automatically established between these views by simultaneously
matching the tensors of a view with those of the remaining views using
a hash table based voting scheme. This results in a graph of relative
transformations used to register the views before they are integrated
into a seamless 3D model. These models and their tensor representations
constitute the model library. During online
recognition, a tensor from the scene is simultaneously matched with
those in the library by casting votes. Similarity measures are
calculated for the model tensors which receive the most votes. The
model with the highest similarity is transformed to the scene and if it
aligns accurately with an object in the scene, that object is declared
as recognized and is segmented. This process is repeated until the
scene is completely segmented. Experiments were performed on real and
synthetic data comprised of 55 models and 610 scenes and an overall
recognition rate of 95 percent was achieved. Comparison with the spin
images
revealed that our algorithm is superior in terms of recognition rate
and efficiency.
Ajmal S. Mian, M. Bennamoun and R. Owens, "3D Model-based Object
Recognition and Segmentation in Cluttered Scenes", to appear in IEEE
Transactions in Pattern Analysis and Machine Intelligence (PAMI), 2006.
[pdf]
Copyright ©
IEEE.
The data we
used in our experiments is also available for download below.
Please cite the above paper if you use the data.
3D Models in PLY format
Chef [download
3.6MB] Parasaurolophus [download
3.6MB] T-rex [download
3.5MB] Chicken [download
2.7MB] Rhino [download
1.6MB]
The above objects were placed in a scene causing
occlusions and clutter. The scene was then scanned with the
Minolta Vivid 910 scanner to get a 2.5D view of the scene. The scenes
are available
for download below. File sizes vary from 1.5 to 2.2 MB.
2.5D Scenes in PLY format (zip files)