Detecting Nonexistent Pedestrians

Chien, Jui-Ting; Chou, Chia-Jung; Chen, Ding-Jie; Chen, Hwann-Tzong

doi:10.1109/iccvw.2017.30

Cited by 13 publications

(21 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Quantitative evaluations. To quantify the realism of the inserted object, an object detector is often used to locate the inserted object [14,19,5]. The premise is that a detector is likely to locate only well-inserted objects since state-of- the-art methods take both of the object and its surrounding context into account.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Inserting Videos Into Videos

Lee

Pfister

Yang

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

In this paper, we introduce a new problem of manipulating a given video by inserting other videos into it. Our main task is, given an object video and a scene video, to insert the object video at a user-specified location in the scene video so that the resulting video looks realistic. We aim to handle different object motions and complex backgrounds without expensive segmentation annotations. As it is difficult to collect training pairs for this problem, we synthesize fake training pairs that can provide helpful supervisory signals when training a neural network with unpaired real data. The proposed network architecture can take both real and fake pairs as input and perform both supervised and unsupervised training in an adversarial learning scheme. To synthesize a realistic video, the network renders each frame based on the current input and previous frames. Within this framework, we observe that injecting noise into previous frames while generating the current frame stabilizes training. We conduct experiments on real-world videos in object tracking and person re-identification benchmark datasets. Experimental results demonstrate that the proposed algorithm is able to synthesize long sequences of realistic videos with a given object video inserted.

show abstract

Section: Resultsmentioning

confidence: 99%

“…To learn both placement and shape of a new object, the method in [5] removes existing objects from the scene using an image in-painting algorithm. Then, a network is trained to recover the existing objects.…”

Section: Related Workmentioning

confidence: 99%

Inserting Videos Into Videos

Lee

Pfister

Yang

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

show abstract

“…Sun and Jacobs [26] learn a representation of context for locating missing objects in an image, such as curb ramps on sidewalks that ought to be present but are not. Like us, Chien et al [5] detect non-existent pedestrians. However, unlike our 3D-aware data collection approach, they use images with labeled pedestrians and remove a subset of them using image inpainting.…”

Section: Related Workmentioning

confidence: 88%

Hidden Footprints: Learning Contextual Walkability from 3D Human Trails

Sun

Averbuch‐Elor

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

Fig. 1: (a) Imagine an urban scene like this one. Where could people walk in such a scene? The answer is not simply 'crosswalk' or 'sidewalk'-scene context plays a significant role. For example, people can be standing on sidewalks and not on crosswalks while waiting for a green light (b, top). A road can be open to walk, or only partially walkable (e.g., when occupied by cows, as in (b, bottom)). (c) We propose to predict where people could walk (highlighted in green) by learning from observing human behavior in weakly annotated data.

show abstract

“…Izadinia et al [40] encoded the scene category, the context-specific appearances of objects and their layouts to learn scene structures. Chien et al [41] built a CNN to predict the probability of observing a pedestrian at some location in image. Wang et al [42] used a variational auto-encoder to extract the scale and deformation of the human pose and thus predict opportunities of interaction in a scene.…”

Section: Context Modeling For Scene Analysismentioning

confidence: 99%

Bounding Boxes Are All We Need: Street View Image Classification via Context Encoding of Detected Buildings

Zhao,

Liu,

Hao

et al. 2020

Preprint

View full text Add to dashboard Cite

Street view images classification aiming at urban land use analysis is difficult because the class labels (e.g., commercial area), are concepts with higher abstract level compared to the ones of general visual tasks (e.g., persons and cars). Therefore, classification models using only visual features often fail to achieve satisfactory performance. In this paper, a novel approach based on a "Detector-Encoder-Classifier" framework is proposed. Instead of using visual features of the whole image directly as common image-level models based on convolutional neural networks (CNNs) do, the proposed framework firstly obtains the bounding boxes of buildings in street view images from a detector. Their contextual information such as the co-occurrence patterns of building classes and their layout are then encoded into metadata by the proposed algorithm "CODING" (Context encOding of Detected buildINGs). Finally, these bounding box metadata are classified by a recurrent neural network (RNN). In addition, we made a dual-labeled dataset named "BEAUTY" (Building dEtection And Urban funcTional-zone portraYing) of 19,070 street view images and 38,857 buildings based on the existing BIC GSV [1]. The dataset can be used not only for street view image classification, but also for multi-class building detection. Experiments on "BEAUTY" show that the proposed approach achieves a 12.65% performance improvement on macro-precision and 12% on macro-recall over image-level CNN based models. Our code and dataset are available at https: //github.com/kyle-one/Context-Encoding-of-Detected-Buildings/ Index Terms-Street view images classification, context encoding, building detection, urban land use classification, urban functional zone, RNN.

show abstract

Detecting Nonexistent Pedestrians

Cited by 13 publications

References 15 publications

Inserting Videos Into Videos

Inserting Videos Into Videos

Hidden Footprints: Learning Contextual Walkability from 3D Human Trails

Bounding Boxes Are All We Need: Street View Image Classification via Context Encoding of Detected Buildings

Contact Info

Product

Resources

About