Probabilistic categorization of kitchen objects in table settings with a composite sensor

ToF cameras for active vision in robotics

“…A classical solution in the area of object modeling is the use of calibrated stereo rigs. Therefore, initial works were devoted to their comparison with [37] Dynamic object detection and classification Color and light independence PMD Hussmann and Liepert [38] Object pose Easy object/background segmentation PMD Guomundsson et al [39] Known object pose estimation Light independent / Absolute scale SR3 Beder et al [40] Surface reconstruction using patchlets ToF easily combines with stereo PMD Fuchs and May [7] Precise surface reconstruction 3D at high rate SR3/O3D100 (Depth) Dellen et al [5] 3D object reconstruction 3D at high rate SR3 (Depth) Foix et al [6] Kuehnle et al [8] Object recognition for grasping 3D allow geometric primitives search SR3 Grundmann et al [41] Collision free object manipulation 3D at high rate SR3 + stereo Reiser and Kubacki [42] Position based visual servoing 3D is simply obtained / No model needed SR3 (Depth) Gachter et al [43] Object part detection for classification 3D at high rate SR3 Shin et al [44] SR2 Klank et al [45] Mobile manipulation Easy table/object segmentation SR4 Marton et al [46] Object categorization ToF easily combines with stereo SR4 + color Nakamura et al [47] Mobile manipulation Easy table segmentation SR4 + color Saxena et al [9] Grasping unknown objects 3D at high rate SR3 + stereo Zhu et al [48] Short range depth maps ToF easily combines with stereo SR3 + stereo Lindner et al [49] Object segmentation for recognition Easy color registration PMD + color camera Fischer et al [50] Occlusion handling in virtual objects 3D at high rate PMD + color camera…”

“…Another advantage is that some object segmentation algorithms can be developed combining cues from both a ToF sensor and a color camera. Using such a combined sensor, Marton et al [46] proposed a probabilistic categorization algorithm for kitchen objects. This work uses a new SR4000 camera.…”

Real-time 3D segmentation of cluttered scenes for robot grasping

“…Various feature extraction methods have been proposed, including 3D-augmented SIFT features [1], [8] and features directly obtained from range images such as a viewpoint feature histogram [3], depth-encoded hough voting [4], point pair features [15], and iterative clustering-estimation [16]. In [18] RGB-stereo, time-of-flight, and thermal cameras are fused to segment kitchen objects in a table-top setup. The addition of other sensor modalities, such as temperature, improves the segmentation accuracy and enables the method to segment shiny and translucent object.…”

Reconstruction and Verification of 3D Object Models for Grasping

“…to find out how probable is that a combinations of fitted models to a cluster is approximating the true 3D structure correctly (employing similar techniques as in [10]). For this, a large number of hand-labeled training sets of correct and incorrect fits is needed, for which the slight variations for the same data introduced by the random sample consensus method works to our advantage.…”