0000000000165821
AUTHOR
Guillaume Allibert
Modality-Guided Subnetwork for Salient Object Detection
Recent RGBD-based models for saliency detection have attracted research attention. The depth clues such as boundary clues, surface normal, shape attribute, etc., contribute to the identification of salient objects with complicated scenarios. However, most RGBD networks require multi-modalities from the input side and feed them separately through a two-stream design, which inevitably results in extra costs on depth sensors and computation. To tackle these inconveniences, we present in this paper a novel fusion design named modality-guided subnetwork (MGSnet). It has the following superior designs: 1) Our model works for both RGB and RGBD data, and dynamically estimating depth if not availabl…
OmniFlowNet: a Perspective Neural Network Adaptation for Optical Flow Estimation in Omnidirectional Images
International audience; Spherical cameras and the latest image processing techniques open up new horizons. In particular, methods based on Convolutional Neural Networks (CNNs) now give excellent results for optical flow estimation on perspective images. However, these approaches are highly dependent on their architectures and training datasets. This paper proposes to benefit from years of improvement in perspective images optical flow estimation and to apply it to omnidirectional ones without training on new datasets. Our network, OmniFlowNet, is built on a CNN specialized in perspective images. Its convolution operation is adapted to be consistent with the equirectangular projection. Teste…
Deep Reinforcement Learning with Omnidirectional Images: application to UAV Navigation in Forests
Deep Reinforcement Learning (DRL) is highly efficient for solving complex tasks such as drone obstacle avoidance using cameras. However, these methods are often limited by the camera perception capabilities. In this paper, we demonstrate that point-goal navigation performances can be improved by using cameras with a wider Field-Of-View (FOV). To this end, we present a DRL solution based on equirectangular images and demonstrates its relevance, especially compared to its perspective version. Several visual modalities are compared: ground truth depth, RGB, and depth directly estimated from these 360°R GB images using Deep Learning methods. Next, we propose a spherical adaptation to take into …
Depth-Adapted CNN for RGB-D cameras
Conventional 2D Convolutional Neural Networks (CNN) extract features from an input image by applying linear filters. These filters compute the spatial coherence by weighting the photometric information on a fixed neighborhood without taking into account the geometric information. We tackle the problem of improving the classical RGB CNN methods by using the depth information provided by the RGB-D cameras. State-of-the-art approaches use depth as an additional channel or image (HHA) or pass from 2D CNN to 3D CNN. This paper proposes a novel and generic procedure to articulate both photometric and geometric information in CNN architecture. The depth data is represented as a 2D offset to adapt …
Robust RGB-D Fusion for Saliency Detection
Efficiently exploiting multi-modal inputs for accurate RGB-D saliency detection is a topic of high interest. Most existing works leverage cross-modal interactions to fuse the two streams of RGB-D for intermediate features' enhancement. In this process, a practical aspect of the low quality of the available depths has not been fully considered yet. In this work, we aim for RGB-D saliency detection that is robust to the low-quality depths which primarily appear in two forms: inaccuracy due to noise and the misalignment to RGB. To this end, we propose a robust RGB-D fusion method that benefits from (1) layer-wise, and (2) trident spatial, attention mechanisms. On the one hand, layer-wise atten…
OMNI-DRL: Learning to Fly in Forests with Omnidirectional Images
Perception is crucial for drone obstacle avoidance in complex, static, and unstructured outdoor environments. However, most navigation solutions based on Deep Reinforcement Learning (DRL) use limited Field-Of-View (FOV) images as input. In this paper, we demonstrate that omnidirectional images improve these methods. Thus, we provide a comparative benchmark of several visual modalities for navigation: ground truth depth, ground truth semantic segmentation, and RGB images. These exhaustive comparisons reveal that it is superior to use an omnidirectional camera to navigate with classical DRL methods. Finally, we show in two different virtual forest environments that adapting the convolution to…