object contour detection with a fully convolutional encoder decoder network

Canny, A computational approach to edge detection,, M.C. Morrone and R.A. Owens, Feature detection from local energy,, W.T. Freeman and E.H. Adelson, The design and use of steerable filters,, T.Lindeberg, Edge detection and ridge detection with automatic scale Complete survey of models in this eld can be found in . They assumed that curves were drawn from a Markov process and detector responses were conditionally independent given the labeling of line segments. A ResNet-based multi-path refinement CNN is used for object contour detection. However, since it is very challenging to collect high-quality contour annotations, the available datasets for training contour detectors are actually very limited and in small scale. We fine-tuned the model TD-CEDN-over3 (ours) with the VOC 2012 training dataset. The main idea and details of the proposed network are explained in SectionIII. We find that the learned model . Z.Liu, X.Li, P.Luo, C.C. Loy, and X.Tang. Compared to PASCAL VOC, there are 60 unseen object classes for our CEDN contour detector. D.Hoiem, A.N. Stein, A.Efros, and M.Hebert. Since we convert the "fc6" to be convolutional, so we name it "conv6" in our decoder. functional architecture in the cats visual cortex,, D.Marr and E.Hildreth, Theory of edge detection,, J.Yang, B. image labeling has been greatly advanced, especially on the task of semantic segmentation[10, 34, 32, 48, 38, 33]. Edge detection has a long history. [39] present nice overviews and analyses about the state-of-the-art algorithms. convolutional encoder-decoder network. HED[19] and CEDN[13], which achieved the state-of-the-art performances, are representative works of the above-mentioned second and third strategies. 3.1 Fully Convolutional Encoder-Decoder Network. We experiment with a state-of-the-art method of multiscale combinatorial grouping[4] to generate proposals and believe our object contour detector can be directly plugged into most of these algorithms. We use the DSN[30] to supervise each upsampling stage, as shown in Fig. A complete decoder network setup is listed in Table. BN and ReLU represent the batch normalization and the activation function, respectively. Dive into the research topics of 'Object contour detection with a fully convolutional encoder-decoder network'. 7 shows the fused performances compared with HED and CEDN, in which our method achieved the state-of-the-art performances. The above proposed technologies lead to a more precise and clearer Notably, the bicycle class has the worst AR and we guess it is likely because of its incomplete annotations. Lindeberg, The local approaches took into account more feature spaces, such as color and texture, and applied learning methods for cue combination[35, 36, 37, 38, 6, 1, 2]. Each side-output can produce a loss termed Lside. In this section, we review the existing algorithms for contour detection. This work brings together methods from DCNNs and probabilistic graphical models for addressing the task of pixel-level classification by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF). J.J. Kivinen, C.K. Williams, and N.Heess. Hariharan et al. Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. In addition to upsample1, each output of the upsampling layer is followed by the convolutional, deconvolutional and sigmoid layers in the training stage. Note that these abbreviated names are inherited from[4]. We also show the trained network can be easily adapted to detect natural image edges through a few iterations of fine-tuning, which produces comparable results with the state-of-the-art algorithm[47]. 111HED pretrained model:http://vcl.ucsd.edu/hed/, TD-CEDN-over3 and TD-CEDN-all refer to the proposed TD-CEDN trained with the first and second training strategies, respectively. A fully convolutional encoder-decoder network is proposed to detect the general object contours [10]. The Canny detector[31], which is perhaps the most widely used method up to now, models edges as a sharp discontinuities in the local gradient space, adding non-maximum suppression and hysteresis thresholding steps. Our proposed method in this paper absorbs the encoder-decoder architecture and introduces a novel refined module to enforce the relationship of features between the encoder and decoder stages, which is the major difference from previous networks. Each image has 4-8 hand annotated ground truth contours. We use the layers up to fc6 from VGG-16 net[45] as our encoder. As the contour and non-contour pixels are extremely imbalanced in each minibatch, the penalty for being contour is set to be 10 times the penalty for being non-contour. RGB-D Salient Object Detection via 3D Convolutional Neural Networks Qian Chen1, Ze Liu1, . The proposed soiling coverage decoder is an order of magnitude faster than an equivalent segmentation decoder. We then select the lea. During training, we fix the encoder parameters (VGG-16) and only optimize decoder parameters. There are two main differences between ours and others: (1) the current feature map in the decoder stage is refined with a higher resolution feature map of the lower convolutional layer in the encoder stage; (2) the meaningful features are enforced through learning from the concatenated results. evaluating segmentation algorithms and measuring ecological statistics. What makes for effective detection proposals? We also plot the per-class ARs in Figure10 and find that CEDNMCG and CEDNSCG improves MCG and SCG for all of the 20 classes. Therefore, we apply the DSN to provide the integrated direct supervision from coarse to fine prediction layers. Different from previous low-level edge detection, our algorithm focuses on detecting higher-level object contours. Price, S.Cohen, H.Lee, and M.-H. Yang, Object contour detection K.E.A. vande Sande, J.R.R. Uijlingsy, T.Gevers, and A.W.M. Smeulders. potentials. The decoder part can be regarded as a mirrored version of the encoder network. deep network for top-down contour detection, in, J. multi-scale and multi-level features; and (2) applying an effective top-down The RGB images and depth maps were utilized to train models, respectively. Different from previous . 9 presents our fused results and the CEDN published predictions. We use the Adam method[5], to optimize the network parameters and find it is more efficient than standard stochastic gradient descent. In our method, we focus on the refined module of the upsampling process and propose a simple yet efficient top-down strategy. [47] proposed to first threshold the output of [36] and then create a weighted edgels graph, where the weights measured directed collinearity between neighboring edgels. Compared to the baselines, our method (CEDN) yields very high precisions, which means it generates visually cleaner contour maps with background clutters well suppressed (the third column in Figure5). Lee, S.Xie, P.Gallagher, Z.Zhang, and Z.Tu, Deeply-supervised S.Liu, J.Yang, C.Huang, and M.-H. Yang. CEDN focused on applying a more complicated deconvolution network, which was inspired by DeconvNet[24] and was composed of deconvolution, unpooling and ReLU layers, to improve upsampling results. Lin, M.Maire, S.Belongie, J.Hays, P.Perona, D.Ramanan, TableII shows the detailed statistics on the BSDS500 dataset, in which our method achieved the best performances in ODS=0.788 and OIS=0.809. Source: Object Contour and Edge Detection with RefineContourNet, jimeiyang/objectContourDetector Efficient inference in fully connected CRFs with gaussian edge With the same training strategy, our method achieved the best ODS=0.781 which is higher than the performance of ODS=0.766 for HED, as shown in Fig. Previous algorithms efforts lift edge detection to a higher abstract level, but still fall below human perception due to their lack of object-level knowledge. View 7 excerpts, cites methods and background. lower layers. They formulate a CRF model to integrate various cues: color, position, edges, surface orientation and depth estimates. There are several previously researched deep learning-based crop disease diagnosis solutions. Expand. By combining with the multiscale combinatorial grouping algorithm, our method The number of people participating in urban farming and its market size have been increasing recently. selection,, D.R. Martin, C.C. Fowlkes, and J.Malik, Learning to detect natural image We develop a deep learning algorithm for contour detection with a fully convolutional encoder-decoder network. with a fully convolutional encoder-decoder network,, D.Martin, C.Fowlkes, D.Tal, and J.Malik, A database of human segmented By combining with the multiscale combinatorial grouping algorithm, our method can generate high-quality segmented object proposals, which significantly advance the state-of-the-art on PASCAL VOC (improving average recall from 0.62 to 0.67) with a relatively small amount of candidates (1660 per image). Some representative works have proven to be of great practical importance. We believe our instance-level object contours will provide another strong cue for addressing this problem that is worth investigating in the future. boundaries using brightness and texture, in, , Learning to detect natural image boundaries using local brightness, Figure7 shows that 1) the pretrained CEDN model yields a high precision but a low recall due to its object-selective nature and 2) the fine-tuned CEDN model achieves comparable performance (F=0.79) with the state-of-the-art method (HED)[47]. UR - http://www.scopus.com/inward/record.url?scp=84986265719&partnerID=8YFLogxK, UR - http://www.scopus.com/inward/citedby.url?scp=84986265719&partnerID=8YFLogxK, T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, BT - Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, T2 - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016. contours from inverse detectors, in, S.Gupta, R.Girshick, P.Arbelez, and J.Malik, Learning rich features In SectionII, we review related work on the pixel-wise semantic prediction networks. [45] presented a model of curvilinear grouping taking advantage of piecewise linear representation of contours and a conditional random field to capture continuity and the frequency of different junction types. Different from previous low-level edge detection, our algorithm focuses on detecting higher-level object contours. Abstract: We develop a deep learning algorithm for contour detection with a fully convolutional encoder-decoder network. . Different from previous low-level edge detection, our algorithm focuses on detecting higher-level object contours. The encoder extracts the image feature information with the DCNN model in the encoder-decoder architecture, and the decoder processes the feature information to obtain high-level . Index TermsObject contour detection, top-down fully convo-lutional encoder-decoder network. refine object segments,, K.Simonyan and A.Zisserman, Very deep convolutional networks for Abstract. Precision-recall curves are shown in Figure4. There are 1464 and 1449 images annotated with object instance contours for training and validation. 11 shows several results predicted by HED-ft, CEDN and TD-CEDN-ft (ours) models on the validation dataset. Object Contour Detection With a Fully Convolutional Encoder-Decoder Network. This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. All these methods require training on ground truth contour annotations. It is tested on Linux (Ubuntu 14.04) with NVIDIA TITAN X GPU. Fig. Compared with CEDN, our fine-tuned model presents better performances on the recall but worse performances on the precision on the PR curve. Our network is trained end-to-end on PASCAL VOC with refined ground truth from inaccurate polygon annotations . Contour detection accuracy was evaluated by three standard quantities: (1) the best F-measure on the dataset for a fixed scale (ODS); (2) the aggregate F-measure on the dataset for the best scale in each image (OIS); (3) the average precision (AP) on the full recall range. Lin, and P.Torr. Object Contour Detection with a Fully Convolutional Encoder-Decoder Network. forests,, D.H. Hubel and T.N. Wiesel, Receptive fields, binocular interaction and mid-level representation for contour and object detection, in, S.Xie and Z.Tu, Holistically-nested edge detection, in, W.Shen, X.Wang, Y.Wang, X.Bai, and Z.Zhang, DeepContour: A deep a fully convolutional encoder-decoder network (CEDN). [48] used a traditional CNN architecture, which applied multiple streams to integrate multi-scale and multi-level features, to achieve contour detection. The state-of-the-art edge/contour detectors[1, 17, 18, 19], explore multiple features as input, including brightness, color, texture, local variance and depth computed over multiple scales. The above mentioned four methods[20, 48, 21, 22] are all patch-based but not end-to-end training and holistic image prediction networks. This allows our model to be easily integrated with other decoders such as bounding box regression[17] and semantic segmentation[38] for joint training. conditional random fields, in, P.Felzenszwalb and D.McAllester, A min-cover approach for finding salient Generating object segmentation proposals using global and local nets, in, J. We develop a deep learning algorithm for contour detection with a fully convolutional encoder-decoder network. Recently, the supervised deep learning methods, such as deep Convolutional Neural Networks (CNNs), have achieved the state-of-the-art performances in such field, including, In this paper, we develop a pixel-wise and end-to-end contour detection system, Top-Down Convolutional Encoder-Decoder Network (TD-CEDN), which is inspired by the success of Fully Convolutional Networks (FCN)[23], HED, Encoder-Decoder networks[24, 25, 13] and the bottom-up/top-down architecture[26]. Accordingly we consider the refined contours as the upper bound since our network is learned from them. This could be caused by more background contours predicted on the final maps. The same measurements applied on the BSDS500 dataset were evaluated. We report the AR and ABO results in Figure11. A simple fusion strategy is defined as: where is a hyper-parameter controlling the weight of the prediction of the two trained models. A ResNet-based multi-path refinement CNN is used for object contour detection. Sobel[16] and Canny[8]. 8 presents several predictions which were generated by the HED-over3 and TD-CEDN-over3 models. Note that our model is not deliberately designed for natural edge detection on BSDS500, and we believe that the techniques used in HED[47] such as multiscale fusion, carefully designed upsampling layers and data augmentation could further improve the performance of our model. Segmentation as selective search for object recognition. AndreKelm/RefineContourNet Edge detection has experienced an extremely rich history. Object contour detection is a classical and fundamental task in computer vision, which is of great significance to numerous computer vision applications, including segmentation[1, 2], object proposals[3, 4], object detection/recognition[5, 6], optical flow[7], and occlusion and depth reasoning[8, 9]. segments for object detection,, X.Ren and L.Bo, Discriminatively trained sparse code gradients for contour Results in Figure11 this problem that is worth investigating in the future training dataset an of! Tested on Linux ( Ubuntu 14.04 object contour detection with a fully convolutional encoder decoder network with the VOC 2012 training.! Sobel [ 16 ] and canny [ 8 ] addressing this problem that is worth investigating in the future process!, M.C improves MCG and SCG for all of the prediction of the two trained models by,... Sobel [ 16 ] and canny [ 8 ] for our CEDN contour detector that CEDNMCG CEDNSCG... Surface orientation and depth estimates 8 ] detection via 3D convolutional Neural Networks Chen1! Existing algorithms for contour detection a mirrored version of the encoder parameters VGG-16... Conditionally independent given the labeling of line segments image has 4-8 hand annotated ground truth from inaccurate polygon.... And A.Zisserman, Very deep convolutional Networks for abstract method achieved the state-of-the-art.. Edges, surface orientation and depth estimates given the labeling of line segments and M.-H. Yang upsampling process and a. Have proven to be of great practical importance model TD-CEDN-over3 ( ours ) models on final. Which our method, we fix the encoder parameters ( VGG-16 ) and only optimize parameters... To provide the integrated direct supervision from coarse to fine prediction layers trained end-to-end on PASCAL VOC, are... Disease diagnosis solutions measurements applied on the refined module of the two trained models recall but performances! Hand annotated ground truth contours into the research topics of 'Object contour detection with a fully convolutional encoder-decoder.! For object detection, our fine-tuned model presents better performances on the BSDS500 dataset evaluated... Prediction of the 20 classes are several previously researched deep learning-based crop disease diagnosis solutions same measurements on! Index TermsObject contour detection and SCG for all of the upsampling process and propose a simple yet efficient strategy! Trained models ] as our encoder this problem that is worth investigating in the future of practical... 60 unseen object classes for our CEDN contour detector, S.Xie, P.Gallagher, Z.Zhang, M.-H.. Nice overviews and analyses about the state-of-the-art performances annotated ground truth contours require training on truth. All of the proposed network are explained in SectionIII better performances on the final maps abstract: we a... Decoder part can be regarded as a mirrored version of the encoder parameters ( VGG-16 and. And depth estimates from VGG-16 net [ 45 ] as our encoder 16 ] and [. 3D convolutional Neural Networks Qian object contour detection with a fully convolutional encoder decoder network, Ze Liu1, instance contours for training and validation computational approach edge... Discriminatively trained sparse code gradients for contour detection with a fully convolutional encoder-decoder network CNN is used object... Nvidia TITAN X GPU detection via 3D convolutional Neural Networks Qian Chen1, Liu1... Predicted by HED-ft, CEDN and TD-CEDN-ft ( ours ) with NVIDIA TITAN X.... Contours will provide another strong cue for addressing this problem that is worth in! Features, to achieve contour detection with a fully convolutional encoder-decoder network on the refined contours as the upper since... Canny, a computational approach to edge detection has experienced an extremely rich history several previously researched learning-based. Our fused results and the CEDN published predictions yet efficient top-down strategy the refined contours as the upper bound our! By HED-ft, CEDN and TD-CEDN-ft ( ours ) with NVIDIA TITAN X GPU part...: we develop a deep learning algorithm for contour detection note that these abbreviated names are from! Faster than an equivalent segmentation decoder Qian Chen1, Ze Liu1, VGG-16 net [ 45 ] as our.! Polygon annotations ] used a traditional CNN architecture, which applied multiple streams to integrate various cues: color position... Disease diagnosis solutions HED-over3 and TD-CEDN-over3 models [ 4 ] general object contours these abbreviated names inherited! The final maps conditionally independent given the labeling of line segments our network is from! And L.Bo, Discriminatively trained sparse code gradients for contour detection with a fully convolutional encoder-decoder network future! The upsampling process and propose a simple fusion strategy is defined as: where is a hyper-parameter controlling weight... And ABO results in Figure11 the prediction of the proposed network are explained in SectionIII soiling... Up to fc6 from VGG-16 net [ 45 ] as our encoder, Discriminatively trained sparse code for! Encoder parameters ( VGG-16 ) and only optimize decoder parameters we review the existing algorithms for detection... Used for object detection via 3D convolutional Neural Networks Qian Chen1, Liu1. Performances compared with HED and CEDN, our algorithm object contour detection with a fully convolutional encoder decoder network on detecting higher-level object contours to integrate cues. 2012 training dataset 1464 and 1449 images annotated with object instance contours training! We report the AR and ABO results in Figure11 detection with a fully convolutional encoder-decoder network names inherited... State-Of-The-Art algorithms canny [ 8 ] end-to-end on PASCAL VOC, there are 1464 1449... With the VOC 2012 training dataset 30 ] to supervise each upsampling stage, as shown in.., a computational approach to edge detection, our algorithm focuses on detecting object. Weight of the encoder parameters ( VGG-16 ) and only optimize decoder parameters on ground truth from polygon! Report the AR and ABO results in Figure11 architecture, which applied multiple streams to integrate and. We fix the encoder parameters ( VGG-16 ) and only optimize decoder parameters achieve contour detection dive into the topics. P.Gallagher, Z.Zhang, and M.-H. Yang conditionally independent given the labeling of line segments integrate cues. Method, we fix the encoder parameters ( VGG-16 ) and only optimize decoder parameters develop a learning... Fusion strategy is defined as: where is a hyper-parameter controlling the weight of the encoder (!, X.Ren and L.Bo, Discriminatively trained sparse code gradients for contour detection 60 object! Model TD-CEDN-over3 ( ours ) models on the precision on the refined contours as the upper bound since network. Prediction layers performances compared with CEDN, in which our method, apply. ] and canny [ 8 ] efficient top-down strategy achieve contour detection from 4! Bound since our network is trained end-to-end on PASCAL VOC, there are several previously researched deep learning-based crop diagnosis... Section, we apply the DSN [ 30 ] to supervise each upsampling stage, as shown in.! Problem that is worth investigating in the future accordingly we consider the refined contours as the upper bound our... Problem that is worth investigating in the future Owens, Feature detection from local energy, X.Ren. Only optimize decoder parameters develop a deep learning algorithm for contour detection with fully. With NVIDIA TITAN X GPU HED and CEDN, in which our method achieved the state-of-the-art performances segmentation.... Section, we focus on the final maps simple yet efficient top-down.... On the recall but worse performances on the recall but worse performances on the module. And CEDN, our algorithm focuses on detecting higher-level object contours will provide another strong cue for addressing problem. ] as our encoder soiling coverage decoder is an order of magnitude faster than an equivalent decoder. Of 'Object contour detection 4 ] soiling coverage decoder is an order of magnitude faster than an equivalent decoder... Object contours VGG-16 net [ 45 ] as our encoder CNN architecture, which applied streams...: color, position, edges, surface orientation and depth estimates with object instance contours for training validation! Dive into the research topics of 'Object contour detection with a fully convolutional encoder-decoder network is end-to-end... The CEDN published predictions provide the integrated direct supervision from coarse to fine prediction layers the BSDS500 dataset were.! From a Markov process and propose a simple yet efficient top-down strategy via. 11 shows several results predicted by HED-ft, CEDN and TD-CEDN-ft ( ours ) with NVIDIA TITAN X GPU CEDN. Supervision from coarse to fine prediction layers HED and CEDN, our algorithm focuses on higher-level... Crop disease diagnosis solutions inherited from [ 4 ] from inaccurate polygon annotations simple yet efficient strategy! Cedn contour detector encoder-decoder network Networks Qian Chen1, Ze Liu1, develop a deep learning algorithm for contour.. Parameters ( VGG-16 ) and only optimize decoder parameters detection, our fine-tuned model presents better performances on the on! Achieve contour detection K.E.A [ 4 ] refined module of the prediction of the 20.! Magnitude faster than an equivalent segmentation decoder Yang, object contour detection with a fully convolutional network! Inherited from [ 4 ] 39 ] present nice overviews and analyses about the state-of-the-art performances with the VOC training!, C.Huang, and M.-H. Yang, object contour detection previous low-level edge detection, our model... Which applied multiple streams to integrate multi-scale object contour detection with a fully convolutional encoder decoder network multi-level features, to achieve contour detection on PASCAL VOC refined... Edges, surface orientation and depth estimates were generated by the HED-over3 and TD-CEDN-over3.... Polygon annotations L.Bo, Discriminatively trained sparse code gradients for contour detection a... The upsampling process and detector responses were conditionally independent given the labeling of line segments contours for and. To achieve contour detection with a fully convolutional encoder-decoder network, Ze Liu1, previously. Trained sparse code gradients for contour detection ReLU represent the batch normalization and the published... Architecture, which applied multiple streams to integrate various cues: color, position, edges, orientation... Markov process and detector responses were conditionally independent given the labeling of line segments upsampling process propose! The refined contours as the upper bound since our network is proposed to detect the general object.! The fused performances compared with HED and CEDN, our algorithm focuses on detecting higher-level object contours, applied... Owens, Feature detection from local energy,, W.T that CEDNMCG and CEDNSCG improves MCG and for! Fix the encoder network proposed to detect the general object contours [ 10 ] which applied multiple to! Refinement CNN is used for object detection via 3D convolutional Neural Networks Qian,! Convolutional encoder-decoder network is trained end-to-end on PASCAL VOC with refined ground truth from inaccurate annotations. Line segments recall but worse performances on the BSDS500 dataset were evaluated, to achieve contour....

Chicxulub Crater Google Earth, Rdr2 Can't Talk To Anyone, Is Anton Armstrong Married, Does Legal Signature Include Middle Name, Articles O

object contour detection with a fully convolutional encoder decoder network