We present HetNet (Multi-level \textbf{Het}erogeneous \textbf{Net}work), a highly efficient mirror detection network. Current mirror detection methods focus more on performance than efficiency, limiting the real-time applications (such as drones). Their lack of efficiency is aroused by the common design of adopting homogeneous modules at different levels, which ignores the difference between different levels of features. In contrast, HetNet detects potential mirror regions initially through low-level understandings (\textit{e.g.}, intensity contrasts) and then combines with high-level understandings (contextual discontinuity for instance) to finalize the predictions. To perform accurate yet efficient mirror detection, HetNet follows an effective architecture that obtains specific information at different stages to detect mirrors. We further propose a multi-orientation intensity-based contrasted module (MIC) and a reflection semantic logical module (RSL), equipped on HetNet, to predict potential mirror regions by low-level understandings and analyze semantic logic in scenarios by high-level understandings, respectively. Compared to the state-of-the-art method, HetNet runs 664$\%$ faster and draws an average performance gain of 8.9$\%$ on MAE, 3.1$\%$ on IoU, and 2.0$\%$ on F-measure on two mirror detection benchmarks.
Pre-training across 3D vision and language remains under development because of limited training data. Recent works attempt to transfer vision-language pre-training models to 3D vision. PointCLIP converts point cloud data to multi-view depth maps, adopting CLIP for shape classification. However, its performance is restricted by the domain gap between rendered depth maps and images, as well as the diversity of depth distributions. To address this issue, we propose CLIP2Point, an image-depth pre-training method by contrastive learning to transfer CLIP to the 3D domain, and adapt it to point cloud classification. We introduce a new depth rendering setting that forms a better visual effect, and then render 52,460 pairs of images and depth maps from ShapeNet for pre-training. The pre-training scheme of CLIP2Point combines cross-modality learning to enforce the depth features for capturing expressive visual and textual features and intra-modality learning to enhance the invariance of depth aggregation. Additionally, we propose a novel Dual-Path Adapter (DPA) module, i.e., a dual-path structure with simplified adapters for few-shot learning. The dual-path structure allows the joint use of CLIP and CLIP2Point, and the simplified adapter can well fit few-shot tasks without post-search. Experimental results show that CLIP2Point is effective in transferring CLIP knowledge to 3D vision. Our CLIP2Point outperforms PointCLIP and other self-supervised 3D networks, achieving state-of-the-art results on zero-shot and few-shot classification.
Glass is very common in our daily life. Existing computer vision systems neglect it and thus may have severe consequences, e.g., a robot may crash into a glass wall. However, sensing the presence of glass is not straightforward. The key challenge is that arbitrary objects/scenes can appear behind the glass. In this paper, we propose an important problem of detecting glass surfaces from a single RGB image. To address this problem, we construct the first large-scale glass detection dataset (GDD) and propose a novel glass detection network, called GDNet-B, which explores abundant contextual cues in a large field-of-view via a novel large-field contextual feature integration (LCFI) module and integrates both high-level and low-level boundary features with a boundary feature enhancement (BFE) module. Extensive experiments demonstrate that our GDNet-B achieves satisfying glass detection results on the images within and beyond the GDD testing set. We further validate the effectiveness and generalization capability of our proposed GDNet-B by applying it to other vision tasks, including mirror segmentation and salient object detection. Finally, we show the potential applications of glass detection and discuss possible future research directions.
Existing deraining methods mainly focus on a single input image. With just a single input image, it is extremely difficult to accurately detect rain streaks, remove rain streaks, and restore rain-free images. Compared with a single 2D image, a light field image (LFI) embeds abundant 3D structure and texture information of the target scene by recording the direction and position of each incident ray via a plenoptic camera, which has emerged as a popular device in the computer vision and graphics research communities. In this paper, we propose a novel network, 4D-MGP-SRRNet, for rain streak removal from an LFI. Our method takes as input all sub-views of a rainy LFI. In order to make full use of the LFI, we adopt 4D convolutional layers to build the proposed rain steak removal network to simultaneously process all sub-views of the LFI. In the proposed network, the rain detection model, MGPDNet, with a novel Multi-scale Self-guided Gaussian Process (MSGP) module is proposed to detect rain streaks from all sub-views of the input LFI. Semi-supervised learning is introduced to accurately detect rain streaks by training on both virtual-world rainy LFIs and real-world rainy LFIs at multiple scales via calculating pseudo ground truth for real-world rain streaks. All sub-views subtracting the predicted rain streaks are then fed into a 4D residual model to estimate depth maps. Finally, all sub-views concatenated with the corresponding rain streaks and fog maps converted from the estimated depth maps are fed into a rainy LFI restoring model that is based on the adversarial recurrent neural network to progressively eliminate rain streaks and recover the rain-free LFI. Extensive quantitative and qualitative evaluations conducted on both synthetic LFIs and real-world LFIs demonstrate the effectiveness of our proposed method.
Existing camouflaged object detection (COD) methods rely heavily on large-scale datasets with pixel-wise annotations. However, due to the ambiguous boundary, it is very time-consuming and labor-intensive to annotate camouflage objects pixel-wisely (which takes ~ 60 minutes per image). In this paper, we propose the first weakly-supervised camouflaged object detection (COD) method, using scribble annotations as supervision. To achieve this, we first construct a scribble-based camouflaged object dataset with 4,040 images and corresponding scribble annotations. It is worth noting that annotating the scribbles used in our dataset takes only ~ 10 seconds per image, which is 360 times faster than per-pixel annotations. However, the network directly using scribble annotations for supervision will fail to localize the boundary of camouflaged objects and tend to have inconsistent predictions since scribble annotations only describe the primary structure of objects without details. To tackle this problem, we propose a novel consistency loss composed of two parts: a reliable cross-view loss to attain reliable consistency over different images, and a soft inside-view loss to maintain consistency inside a single prediction map. Besides, we observe that humans use semantic information to segment regions near boundaries of camouflaged objects. Therefore, we design a feature-guided loss, which includes visual features directly extracted from images and semantically significant features captured by models. Moreover, we propose a novel network that detects camouflaged objects by scribble learning on structural information and semantic relations. Experimental results show that our model outperforms relevant state-of-the-art methods on three COD benchmarks with an average improvement of 11.0% on MAE, 3.2% on S-measure, 2.5% on E-measure and 4.4% on weighted F-measure.
Mirror detection aims to identify the mirror regions in the given input image. Existing works mainly focus on integrating the semantic features and structural features to mine the similarity and discontinuity between mirror and non-mirror regions, or introducing depth information to help analyze the existence of mirrors. In this work, we observe that a real object typically forms a loose symmetry relationship with its corresponding reflection in the mirror, which is beneficial in distinguishing mirrors from real objects. Based on this observation, we propose a dual-path Symmetry-Aware Transformer-based mirror detection Network (SATNet), which includes two novel modules: Symmetry-Aware Attention Module (SAAM) and Contrast and Fusion Decoder Module (CFDM). Specifically, we first introduce the transformer backbone to model global information aggregation in images, extracting multi-scale features in two paths. We then feed the high-level dual-path features to SAAMs to capture the symmetry relations. Finally, we fuse the dual-path features and refine our prediction maps progressively with CFDMs to obtain the final mirror mask. Experimental results show that SATNet outperforms both RGB and RGB-D mirror detection methods on all available mirror detection datasets.
Recent works on image harmonization solve the problem as a pixel-wise image translation task via large autoencoders. They have unsatisfactory performances and slow inference speeds when dealing with high-resolution images. In this work, we observe that adjusting the input arguments of basic image filters, e.g., brightness and contrast, is sufficient for humans to produce realistic images from the composite ones. Hence, we frame image harmonization as an image-level regression problem to learn the arguments of the filters that humans use for the task. We present a Harmonizer framework for image harmonization. Unlike prior methods that are based on black-box autoencoders, Harmonizer contains a neural network for filter argument prediction and several white-box filters (based on the predicted arguments) for image harmonization. We also introduce a cascade regressor and a dynamic loss strategy for Harmonizer to learn filter arguments more stably and precisely. Since our network only outputs image-level arguments and the filters we used are efficient, Harmonizer is much lighter and faster than existing methods. Comprehensive experiments demonstrate that Harmonizer surpasses existing methods notably, especially with high-resolution inputs. Finally, we apply Harmonizer to video harmonization, which achieves consistent results across frames and 56 fps at 1080P resolution. Code and models are available at: https://github.com/ZHKKKe/Harmonizer.
Glass surfaces are becoming increasingly ubiquitous as modern buildings tend to use a lot of glass panels. This however poses substantial challenges on the operations of autonomous systems such as robots, self-driving cars and drones, as the glass panels can become transparent obstacles to the navigation.Existing works attempt to exploit various cues, including glass boundary context or reflections, as a prior. However, they are all based on input RGB images.We observe that the transmission of 3D depth sensor light through glass surfaces often produces blank regions in the depth maps, which can offer additional insights to complement the RGB image features for glass surface detection. In this paper, we propose a novel framework for glass surface detection by incorporating RGB-D information, with two novel modules: (1) a cross-modal context mining (CCM) module to adaptively learn individual and mutual context features from RGB and depth information, and (2) a depth-missing aware attention (DAA) module to explicitly exploit spatial locations where missing depths occur to help detect the presence of glass surfaces. In addition, we propose a large-scale RGB-D glass surface detection dataset, called \textit{RGB-D GSD}, for RGB-D glass surface detection. Our dataset comprises 3,009 real-world RGB-D glass surface images with precise annotations. Extensive experimental results show that our proposed model outperforms state-of-the-art methods.
Salient Object Ranking (SOR) involves ranking the degree of saliency of multiple salient objects in an input image. Most recently, a method is proposed for ranking salient objects in an input video based on a predicted fixation map. It relies solely on the density of the fixations within the salient objects to infer their saliency ranks, which is incompatible with human perception of saliency ranking. In this work, we propose to explicitly learn the spatial and temporal relations between different salient objects to produce the saliency ranks. To this end, we propose an end-to-end method for video salient object ranking (VSOR), with two novel modules: an intra-frame adaptive relation (IAR) module to learn the spatial relation among the salient objects in the same frame locally and globally, and an inter-frame dynamic relation (IDR) module to model the temporal relation of saliency across different frames. In addition, to address the limited video types (just sports and movies) and scene diversity in the existing VSOR dataset, we propose a new dataset that covers different video types and diverse scenes on a large scale. Experimental results demonstrate that our method outperforms state-of-the-art methods in relevant fields. We will make the source code and our proposed dataset available.
The saliency ranking task is recently proposed to study the visual behavior that humans would typically shift their attention over different objects of a scene based on their degrees of saliency. Existing approaches focus on learning either object-object or object-scene relations. Such a strategy follows the idea of object-based attention in Psychology, but it tends to favor those objects with strong semantics (e.g., humans), resulting in unrealistic saliency ranking. We observe that spatial attention works concurrently with object-based attention in the human visual recognition system. During the recognition process, the human spatial attention mechanism would move, engage, and disengage from region to region (i.e., context to context). This inspires us to model the region-level interactions, in addition to the object-level reasoning, for saliency ranking. To this end, we propose a novel bi-directional method to unify spatial attention and object-based attention for saliency ranking. Our model includes two novel modules: (1) a selective object saliency (SOS) module that models objectbased attention via inferring the semantic representation of the salient object, and (2) an object-context-object relation (OCOR) module that allocates saliency ranks to objects by jointly modeling the object-context and context-object interactions of the salient objects. Extensive experiments show that our approach outperforms existing state-of-theart methods. Our code and pretrained model are available at https://github.com/GrassBro/OCOR.