Global Hypothesis Generation for 6D Object Pose Estimation

By Frank Michel, Alexander Kirillov, Eric Brachmann, Alexander Krull, Stefan Gumhold, Bogdan Savchynskyy, Carsten Rother. 2016

122 citations recorded by the source corpus.

This paper addresses the task of estimating the 6D-pose of a known 3D object from a single RGB-D image. Most modern approaches solve this task in three steps: i) compute local features, ii) generate a pool of pose-hypotheses, iii) select and refine a pose from the pool. This work focuses on the second step. While all existing approaches generate the hypotheses pool via local reasoning, e.g. RANSAC or Hough-Voting, we are the first to show that global reasoning is beneficial at this stage. In particular, we formulate a novel fully-connected Conditional Random Field (CRF) that outputs a very small number of pose-hypotheses. Despite the potential functions of the CRF being non-Gaussian, we give a new, efficient two-step optimization procedure, with some guarantees for optimality. We utilize our global hypotheses generation procedure to produce results that exceed state-of-the-art for the challenging Occluded Object Dataset.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, 3D Vision & Pose

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and Applications (2022, background): “One highly related direction is to model and estimate 3D hand joints [5,44,45,53,60,63,66,72,78,79,80,85,3] and mesh [26,27,51,55,75,83,46], object pose [14,28,29,35,43,57,64,71,76,81], or both [6,17,18,19,69].”
PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes (2017, methodology): “Since the state-of-the-art 6D pose estimation methods mostly rely on regressing image pixels to 3D object coordinates [7], [11], [25], we implement a variation of our network for 3D object coordinate regression for comparison.”
6-DoF object pose from semantic keypoints (2017, methodology): “An alternative to sparse discriminative keypoints is offered by dense methods [9], [10], [11], [12], where every pixel or patch is voting for the object pose.”
BOP: Benchmark for 6D Object Pose Estimation (2018, background): “[34,24,18,2,36,21,27,25], but it is unclear which methods perform well and in which scenarios.”
Reconstructing Hand-Object Interactions in the Wild (2020, background): “On the other hand, some approaches [24, 28, 53, 57, 45, 39] assume known 3D object model and focus on 6DOF object pose prediction.”
DPOD: 6D Pose Object Detector and Refiner (2019, background): “Detecting 3D objects and estimating their 6D pose has been addressed in many works in the past, but the majority of them used depth or RGB-D cameras [2, 5, 16, 17, 23, 30, 4].”
PVN3D: A Deep Point-Wise 3D Keypoints Voting Network for 6DoF Pose Estimation (2019, methodology): “They either use random forest [3, 32] or CNNs [23, 9, 27, 35, 51] to extract feature and predict the corresponding 3D object coordinates for each pixel and then vote for the ﬁnal pose results.”
Real-Time Seamless Single Shot 6D Object Pose Prediction (2017, background): “The advent of commodity depth cameras has spawned many RGB-D object pose estimation methods [1, 3, 4, 11, 14, 23, 32, 39].”
Geometric Image Synthesis (2018, methodology): “ide, we use the object images from our network to also augment them on NYU dataset scenes. To evaluate the performance of Mask R-CNN, we test it on Linemod-Occluded dataset, proposed by Michel et al. [41] where a sequence of real images picked from Linemod dataset are annotated with instance masks o”
DeepIM: Deep Iterative Matching for 6D Pose Estimation (2018, background): “A common strategy is to estimate an initial pose of an object based on the color image, and then refine the pose using depth-based local refinement algorithms such as ICP (Hinterstoisser et al., 2012b; Michel et al., 2017; Zeng et al., 2017).”
PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation (2018, methodology): “[2, 25] use a random forest to predict 3D object coordinates for each pixel and produce 2D-3D correspondence hypotheses using geometric constraints.”
Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation (2020, background): “[2,12,17] regress the corresponding 3D object coordinate for each foreground pixel.”
Segmentation-Driven 6D Object Pose Estimation (2018, methodology): “In this paper, we focus on 6D object pose estimation from RGB images, without access to a depth map, unlike in RGBD-based methods [12, 2, 3, 28].”