Diverse M-Best Solutions in Markov Random Fields

Batra, Dhruv; Yadollahpour, Payman; Guzman-Rivera, Abner; Shakhnarovich, Gregory

doi:10.1007/978-3-642-33715-4_1

Cited by 110 publications

(160 citation statements)

References 31 publications

Supporting

Mentioning

155

Contrasting

Order By: Relevance

“…Another potential of our algorithm is that since we decompose the process into subproblems, for each stage we can use any proper model for specific tasks. For example, we can use other methods to produce more diverse multiple hypotheses, such as [2,15]. For the categorization, we only use the best positive samples for training, but during the inference, the segmentation results from test images are usually not as good as training ones.…”

Section: Discussionmentioning

confidence: 99%

Decomposed Learning for Joint Segmentation and Categorization

Tsai

Yang

2013

Procedings of the British Machine Vision Conference 2013

View full text Add to dashboard Cite

We present a learning algorithm for joint object segmentation and categorization that decomposes the original problem into two sub-tasks and admits their bidirectional interaction. In the first stage, in order to decompose output space, we train category-specific segmentation models to generate figure-ground hypotheses. In the second stage, by taking advantage of object figure-ground information, we train a multi-class segment-based categorization model to determine the object class. A re-ranking strategy is then applied to classified segments to obtain the final category-level segmentation results. Experiments on the Graz-02 and Caltech-101 datasets show that the proposed algorithm performs favorably against the state-of-the-art methods.

show abstract

Section: Discussionmentioning

confidence: 99%

Decomposed Learning for Joint Segmentation and Categorization

Tsai

Yang

2013

Procedings of the British Machine Vision Conference 2013

View full text Add to dashboard Cite

show abstract

“…Hence, diverse solutions are preferred a over single solution. Inspired by [11], we obtain M -best solutions instead of one map solution. This is done for all the selected candidate words from the previous stage individually.…”

Section: Diversity Preserving Inferencementioning

confidence: 99%

“…We begin by generating a set of candidate words with M-best diverse solutions [11]. With these potential solutions, we refine the large lexicon by removing words from it with a large edit distance to any of the candidates, and then recompute the M-best diverse solutions.…”

Section: Introductionmentioning

confidence: 99%

Scene Text Recognition and Retrieval for Large Lexicons

Roy

Mishra

Alahari

et al. 2015

Computer Vision – ACCV 2014

View full text Add to dashboard Cite

Abstract. In this paper we propose a framework for recognition and retrieval tasks in the context of scene text images. In contrast to many of the recent works, we focus on the case where an image-specific list of words, known as the small lexicon setting, is unavailable. We present a conditional random field model defined on potential character locations and the interactions between them. Observing that the interaction potentials computed in the large lexicon setting are less effective than in the case of a small lexicon, we propose an iterative method, which alternates between finding the most likely solution and refining the interaction potentials. We evaluate our method on public datasets and show that it improves over baseline and state-of-the-art approaches. For example, we obtain nearly 15% improvement in recognition accuracy and precision for our retrieval task over baseline methods on the IIIT-5K word dataset, with a large lexicon containing 0.5 million words.

show abstract

“…This is a particular instance of the general problem proposed in [4]. As in our case ∆ G depends only on the part location, its score can be added directly to the data term of Eq.…”

Section: Multiple Hypothesesmentioning

confidence: 99%

An Elastic Deformation Field Model for Object Detection and Tracking

et al. 2014

View full text Add to dashboard Cite

Deformable Parts Models (DPM) are the current stateof-the-art for object detection. Nevertheless they seem sub-optimal in the representation of deformations. Object deformations are often continuous and not confined to big parts. Therefore we propose to replace the DPM star model based on big parts by a deformation field. This consists of a grid of small parts connected with pairwise constraints which can better handle continuous deformations. The naive application of this model for object detection would consist of a bounded sliding window approach: for each possible location of the image the best part configuration within a limited bound around this location is found. This is computationally very expensive. Instead, we propose a different inference procedure, where an iterative image-level search finds the best object hypothesis. We show that this approach is faster than bounded sliding windows yet produces comparable accuracy. Experiments further show that the deformation field can better approximate real object deformations and therefore, for certain classes, produces even better detection accuracy than state-of-the-art DPM. Finally, the same approach is adapted to model-free tracking, showing improved accuracy also in this case.

show abstract

Diverse M-Best Solutions in Markov Random Fields

Cited by 110 publications

References 31 publications

Decomposed Learning for Joint Segmentation and Categorization

Decomposed Learning for Joint Segmentation and Categorization

Scene Text Recognition and Retrieval for Large Lexicons

An Elastic Deformation Field Model for Object Detection and Tracking

Contact Info

Product

Resources

About