Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. 
The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

In this work, we describe a novel method for creating 3D models of persons freely moving in front of a consumer depth sensor and we show how they can be used for long-term person re-identification. For overcoming the problem of the different poses a person can assume, we exploit the information provided by skeletal tracking algorithms for warping every point cloud frame to a standard pose in real time. Then, the warped point clouds are merged together to compose the model. Re-identification is performed by matching body shapes in terms of whole point clouds warped to a standard pose with the described method. We compare this technique with a classification method based on a descriptor of skeleton features and with a mixed approach which exploits both skeleton and shape features. We report experiments on two datasets we acquired for RGB-D re-identification which use different skeletal tracking algorithms and which are made publicly available to foster research in this new research branch.

3D reconstruction of freely moving persons for re-identification with a depth sensor

The state-of-the art in visual object retrieval from large databases allows to search millions of images on the object level Recently, complementary works have proposed systems to crawl large object databases from community photo collections on the Internet We combine these two lines of work to a large-scale system for auto-annotation of holiday snaps The resulting method allows for automatic labeling objects such as landmark buildings, scenes, pieces of art etc at the object level in a fully automatic manner The labeling is multi-modal and consists of textual tags, geographic location, and related content on the Internet Furthermore, the efficiency of the retrieval process is optimized by creating more compact and precise indices for visual vocabularies using background information obtained in the crawling stage of the system We demonstrate the scalability and precision of the proposed method by conducting experiments on millions of images downloaded from community photo collections on the Internet

I know what you did last summer: object-level auto-annotation of holiday snaps

This year alone has seen unprecedented leaps in the area of learning-based image translation, namely CycleGAN, by Zhu et al. But experiments so far have been tailored to merely two domains at a time, and scaling them to more would require an quadratic number of models to be trained. And with two-domain models taking days to train on current hardware, the number of domains quickly becomes limited by the time and resources required to process them. In this paper, we propose a multi-component image translation model and training scheme which scales linearly - both in resource consumption and time required - with the number of domains. We demonstrate its capabilities on a dataset of paintings by 14 different artists and on images of the four different seasons in the Alps. Note that 14 data groups would need (14 choose 2) = 91 different CycleGAN models: a total of 182 generator/discriminator pairs; whereas our model requires only 14 generator/discriminator pairs.

ComboGAN: Unrestrained Scalability for Image Domain Translation

The paper describes a fast system for appearance based image recognition. It uses local invariant descriptors and efficient nearest neighbor search. First, local affine invariant regions are found nested at multiscale intensity extremas. These regions are characterized by nine generalized color moment invariants. An efficient novel method called HPAT (hyper-polyhedron with adaptive threshold) is introduced for efficient localization of the nearest neighbor in feature space.

The invariants make the method robust against changing illumination and viewpoint. The locality helps to resolve occlusions. The proposed indexing method overcomes the drawbacks of most binary tree-like indexing techniques, namely the high complexity in high dimensional data sets and the boundary problem. The database representation is very compact and the retrieval close to realtime on a standard PC. The performance of the proposed method is demonstrated on a public database containing 1005 images of urban scenes. Experiments with an image database containing objects are also presented.

HPAT indexing for fast object/scene recognition based on local appearance

Current state-of-the-art trackers rely only on a target appearance model in order to localize the object in each frame. Such approaches are however prone to fail in case of e.g. fast appearance changes or presence of distractor objects, where a target appearance model alone is insufficient for robust tracking. Having the knowledge about the presence and locations of other objects in the surrounding scene can be highly beneficial in such cases. This scene information can be propagated through the sequence and used to, for instance, explicitly avoid distractor objects and eliminate target candidate regions.

/pdf/know-your-surroundings-exploiting-scene-information-for-1f3epcmbqa.pdf

Luc Van Gool

Papers

3D reconstruction of freely moving persons for re-identification with a depth sensor

I know what you did last summer: object-level auto-annotation of holiday snaps

ComboGAN: Unrestrained Scalability for Image Domain Translation

HPAT indexing for fast object/scene recognition based on local appearance

Know Your Surroundings: Exploiting Scene Information for Object Tracking