Nonmyopic View Planning for Active Object Classification and Pose Estimation

Atanasov, Nikolay; Sankaran, Bharath; Ny, Jérôme Le; Pappas, George J.; Daniilidis, Kostas

doi:10.1109/tro.2014.2320795

Cited by 86 publications

(63 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However this multi-view problem is intrinsically 3D in nature. Atanasov et al, [1,2] implement the idea in real world robots, but they assume that there is only one object associated with each class reducing their problem to instance-level recognition with no intra-class variance. Similar to [9], we use mutual information to decide the NBV.…”

Section: Related Workmentioning

confidence: 99%

3D ShapeNets: A deep representation for volumetric shapes

Song

Khosla

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

1,646

440

View full text Add to dashboard Cite

Abstract3D shape is a crucial but heavily underutilized cue in today's computer vision systems, mostly due to the lack of a good generic shape representation. With the recent availability of inexpensive 2.5D depth sensors (e.g. Microsoft Kinect), it is becoming increasingly important to have a powerful 3D shape representation in the loop. Apart from category recognition, recovering full 3D shapes from viewbased 2.5D depth maps is also a critical part of visual understanding. To this end, we propose to represent a geometric 3D shape as a probability distribution of binary variables on a 3D voxel grid, using a Convolutional Deep Belief Network. Our model, 3D ShapeNets, learns the distribution of complex 3D shapes across different object categories and arbitrary poses from raw CAD data, and discovers hierarchical compositional part representations automatically. It naturally supports joint object recognition and shape completion from 2.5D depth maps, and it enables active object recognition through view planning. To train our 3D deep learning model, we construct ModelNet -a large-scale 3D CAD model dataset. Extensive experiments show that our 3D deep representation enables significant performance improvement over the-state-of-the-arts in a variety of tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

3D ShapeNets: A deep representation for volumetric shapes

Song

Khosla

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

1,646

440

View full text Add to dashboard Cite

show abstract

“…Scaling the problem up to larger environments results in location-dependent action costs, and therefore performance is significantly improved by planning sequences of viewpoints over longer planning horizons. Some approaches have been proposed for planning sequences of locations [5], [6], [7], but the formulations have been limited to restricted cases, such as a single object or a constrained action space. Little attention has been given to continuous candidate viewpoint regions or multi-robot planning for active perception in object recognition tasks.…”

Section: Related Workmentioning

confidence: 99%

“…The informativeness of observations, and therefore the performance of perception algorithms, can be improved by judiciously selecting observation locations [4]. Performance can be significantly improved by using longer planning horizons [5], [6], [7], jointly planning for multiple robots [8], [9], [10], [11] and considering larger sets of candidate sensing locations. However, current planning algorithms with these properties are often too computationally expensive for practical use in large scale and more complex active perception tasks; we propose a self-organising map algorithm as a solution to bridge this gap.…”

Section: Introductionmentioning

confidence: 99%

Multi-robot path planning for budgeted active perception with self-organising maps

Best

Faigl

2016

2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Abstract-We propose a self-organising map (SOM) algorithm as a solution to a new multi-goal path planning problem for active perception and data collection tasks. We optimise paths for a multi-robot team that aims to maximally observe a set of nodes in the environment. The selected nodes are observed by visiting associated viewpoint regions defined by a sensor model. The key problem characteristics are that the viewpoint regions are overlapping polygonal continuous regions, each node has an observation reward, and the robots are constrained by travel budgets. The SOM algorithm jointly selects and allocates nodes to the robots and finds favourable sequences of sensing locations. The algorithm has polynomial-bounded runtime independent of the number of robots. We demonstrate feasibility for the active perception task of observing a set of 3D objects. The viewpoint regions consider sensing ranges and selfocclusions, and the rewards are measured as discriminability in the ensemble of shape functions feature space. Simulations were performed using a 3D point cloud dataset from a real robot in a large outdoor environment. Our results show the proposed methods enable multi-robot planning for budgeted active perception tasks with continuous sets of candidate viewpoints and long planning horizons.

show abstract

“…Most of these methods rely on uncertainty about object label, and use greedy best next action selection [7] at the test time to decrease label probability entropy. A few method aim for optimal action selection at the test time using *This work was not supported by any organization 1 [2] or Monte Carlo planning [25]. However these methods require a model of the object, and are computationally heavy at the test time.…”

Section: Introductionmentioning

confidence: 99%

Belief tree search for active object recognition

Malmir

Cottrell

2017

2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Active Object Recognition (AOR) has been approached as an unsupervised learning problem, in which optimal trajectories for object inspection are not known and are to be discovered by reducing label uncertainty measures or training with reinforcement learning. Such approaches have no guarantees of the quality of their solution. In this paper, we treat AOR as a Partially Observable Markov Decision Process (POMDP) and find near-optimal policies on training data using Belief Tree Search (BTS) on the corresponding belief Markov Decision Process (MDP). AOR then reduces to the problem of knowledge transfer from near-optimal policies on training set to the test set. We train a Long Short Term Memory (LSTM) network to predict the best next action on the training set rollouts. We sho that the proposed AOR method generalizes well to novel views of familiar objects and also to novel objects. We compare this supervised scheme against guided policy search, and find that the LSTM network reaches higher recognition accuracy compared to the guided policy method. We further look into optimizing the observation function to increase the total collected reward of optimal policy. In AOR, the observation function is known only approximately. We propose a gradient-based method update to this approximate observation function to increase the total reward of any policy. We show that by optimizing the observation function and retraining the supervised LSTM network, the AOR performance on the test set improves significantly.

show abstract

Nonmyopic View Planning for Active Object Classification and Pose Estimation

Cited by 86 publications

References 32 publications

3D ShapeNets: A deep representation for volumetric shapes

3D ShapeNets: A deep representation for volumetric shapes

Multi-robot path planning for budgeted active perception with self-organising maps

Belief tree search for active object recognition

Contact Info

Product

Resources

About