Large-Scale FPGA-Based Convolutional Networks

Farabet, Clément; LeCun, Yann; Kavukcuoglu, Koray; Martini, Berin; Akselrod, Polina; Talay, Selcuk; Culurciello, Eugenio

doi:10.1017/cbo9781139042918.020

Cited by 99 publications

(38 citation statements)

References 32 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…ConvNets are easily amenable to efficient hardware implementations in chips or field-programmable gate arrays 66,67 . A number of companies such as NVIDIA, Mobileye, Intel, Qualcomm and Samsung are developing ConvNet chips to enable real-time vision applications in smartphones, cameras, robots and self-driving cars.…”

Section: Image Understanding With Deep Convolutional Networkmentioning

confidence: 99%

Deep learning

LeCun

Bengio

Hinton

2015

Nature

Self Cite

64,001

35,354

View full text Add to dashboard Cite

Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many other domains such as drug discovery and genomics. Deep learning discovers intricate structure in large data sets by using the backpropagation algorithm to indicate how a machine should change its internal parameters that are used to compute the representation in each layer from the representation in the previous layer. Deep convolutional nets have brought about breakthroughs in processing images, video, speech and audio, whereas recurrent nets have shone light on sequential data such as text and speech.

show abstract

Section: Image Understanding With Deep Convolutional Networkmentioning

confidence: 99%

Deep learning

LeCun

Bengio

Hinton

2015

Nature

Self Cite

64,001

35,354

View full text Add to dashboard Cite

show abstract

“…has more diverse layer types and hence quantization is more challenging. The work of [10] uses a directly quantized CNN. However it does not provide a retraining mechanism with low precision weights.…”

Section: Introductionmentioning

confidence: 99%

Fixed point optimization of deep convolutional neural networks for object recognition

Anwar

Hwang

Sung

2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

186

103

View full text Add to dashboard Cite

Deep convolutional neural networks have shown promising results in image and speech recognition applications. The learning capability of the network improves with increasing depth and size of each layer. However this capability comes at the cost of increased computational complexity. Thus reduction in hardware complexity and faster classification are highly desired. This work proposes an optimization method for fixed point deep convolutional neural networks. The parameters of a pre-trained high precision network are first directly quantized using L2 error minimization. We quantize each layer one by one, while other layers keep computation with high precision, to know the layer-wise sensitivity on word-length reduction. Then the network is retrained with quantized weights. Two examples on object recognition, MNIST and CIFAR-10, are presented. Our results indicate that quantization induces sparsity in the network which reduces the effective number of network parameters and improves generalization. This work reduces the required memory storage by a factor of 1/10 and achieves better classification results than the high precision networks.Index Terms-convolutional neural network, quantization, word length optimization, sparsity

show abstract

“…Our implementation takes approximately 5 seconds to output the features of a 320 × 240 image and approximately 5 seconds to estimate the optimum labeling. Further, our approach is highly parallelizable and specially suitable for FPGA-based processors [15]. From these results, we can conclude that the proposed CNN-based algorithm provide promising semantic road scene segmentation in a single image.…”

Section: Resultsmentioning

confidence: 64%

Semantic Road Segmentation via Multi-scale Ensembles of Learned Features

Álvarez

LeCun

Gevers

et al. 2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

Self Cite

View full text Add to dashboard Cite

Abstract. Semantic segmentation refers to the process of assigning an object label (e.g., building, road, sidewalk, car, pedestrian) to every pixel in an image. Common approaches formulate the task as a random field labeling problem modeling the interactions between labels by combining local and contextual features such as color, depth, edges, SIFT or HoG. These models are trained to maximize the likelihood of the correct classification given a training set. However, these approaches rely on hand-designed features (e.g., texture, SIFT or HoG) and a higher computational time required in the inference process.Therefore, in this paper, we focus on estimating the unary potentials of a conditional random field via ensembles of learned features. We propose an algorithm based on convolutional neural networks to learn local features from training data at different scales and resolutions. Then, diversification between these features is exploited using a weighted linear combination. Experiments on a publicly available database show the effectiveness of the proposed method to perform semantic road scene segmentation in still images. The algorithm outperforms appearance based methods and its performance is similar compared to state-of-the-art methods using other sources of information such as depth, motion or stereo.

show abstract

Large-Scale FPGA-Based Convolutional Networks

Cited by 99 publications

References 32 publications

Deep learning

Deep learning

Fixed point optimization of deep convolutional neural networks for object recognition

Semantic Road Segmentation via Multi-scale Ensembles of Learned Features

Contact Info

Product

Resources

About