This paper reviews the NTIRE 2024 challenge on image super-resolution ($\times$4), highlighting the solutions proposed and the outcomes obtained. The challenge involves generating corresponding high-resolution (HR) images, magnified by a factor of four, from low-resolution (LR) inputs using prior information. The LR images originate from bicubic downsampling degradation. The aim of the challenge is to obtain designs/solutions with the most advanced SR performance, with no constraints on computational resources (e.g., model size and FLOPs) or training data. The track of this challenge assesses performance with the PSNR metric on the DIV2K testing dataset. The competition attracted 199 registrants, with 20 teams submitting valid entries. This collective endeavour not only pushes the boundaries of performance in single-image SR but also offers a comprehensive overview of current trends in this field.
In this work we address the problem of rain streak removal with RAW images. The general approach is firstly processing RAW data into RGB images and removing rain streak with RGB images. Actually the original information of rain in RAW images is affected by image signal processing (ISP) pipelines including none-linear algorithms, unexpected noise, artifacts and so on. It gains more benefit to directly remove rain in RAW data before being processed into RGB format. To solve this problem, we propose a joint solution for rain removal and RAW processing to obtain clean color images from rainy RAW image. To be specific, we generate rainy RAW data by converting color rain streak into RAW space and design simple but efficient RAW processing algorithms to synthesize both rainy and clean color images. The rainy color images are used as reference to help color corrections. Different backbones show that our method conduct a better result compared with several other state-of-the-art deraining methods focused on color image. In addition, the proposed network generalizes well to other cameras beyond our selected RAW dataset. Finally, we give the result tested on images processed by different ISP pipelines to show the generalization performance of our model is better compared with methods on color images.
Data augmentation via back-translation is common when pretraining Vision-and-Language Navigation (VLN) models, even though the generated instructions are noisy. But: does that noise matter? We find that nonsensical or irrelevant language instructions during pretraining can have little effect on downstream performance for both HAMT and VLN-BERT on R2R, and is still better than only using clean, human data. To underscore these results, we concoct an efficient augmentation method, Unigram + Object, which generates nonsensical instructions that nonetheless improve downstream performance. Our findings suggest that what matters for VLN R2R pretraining is the quantity of visual trajectories, not the quality of instructions.
Sampling-based path planning algorithms suffer from heavy reliance on uniform sampling, which accounts for unreliable and time-consuming performance, especially in complex environments. Recently, neural-network-driven methods predict regions as sampling domains to realize a non-uniform sampling and reduce calculation time. However, the accuracy of region prediction hinders further improvement. We propose a sampling-based algorithm, abbreviated to Region Prediction Neural Network RRT* (RPNN-RRT*), to rapidly obtain the optimal path based on a high-accuracy region prediction. First, we implement a region prediction neural network (RPNN), to predict accurate regions for the RPNN-RRT*. A full-layer channel-wise attention module is employed to enhance the feature fusion in the concatenation between the encoder and decoder. Moreover, a three-level hierarchy loss is designed to learn the pixel-wise, map-wise, and patch-wise features. A dataset, named Complex Environment Motion Planning, is established to test the performance in complex environments. Ablation studies and test results show that a high accuracy of 89.13% is achieved by the RPNN for region prediction, compared with other region prediction models. In addition, the RPNN-RRT* performs in different complex scenarios, demonstrating significant and reliable superiority in terms of the calculation time, sampling efficiency, and success rate for optimal path planning.
Multi-goal path finding (MGPF) aims to find a closed and collision-free path to visit a sequence of goals orderly. As a physical travelling salesman problem, an undirected complete graph with accurate weights is crucial for determining the visiting order. Lack of prior knowledge of local paths between vertices poses challenges in meeting the optimality and efficiency requirements of algorithms. In this study, a multi-task learning model designated Prior Knowledge Extraction (PKE), is designed to estimate the local path length between pairwise vertices as the weights of the graph. Simultaneously, a promising region and a guideline are predicted as heuristics for the path-finding process. Utilizing the outputs of the PKE model, a variant of Rapidly-exploring Random Tree (RRT) is proposed known as PKE-RRT. It effectively tackles the MGPF problem by a local planner incorporating a prioritized visiting order, which is obtained from the complete graph. Furthermore, the predicted region and guideline facilitate efficient exploration of the tree structure, enabling the algorithm to rapidly provide a sub-optimal solution. Extensive numerical experiments demonstrate the outstanding performance of the PKE-RRT for the MGPF problem with a different number of goals, in terms of calculation time, path cost, sample number, and success rate.
In this paper, we propose a novel end-to-end approach for solving the multi-goal path planning problem in obstacle environments. Our proposed model, called S&Reg, integrates multi-task learning networks with a TSP solver and a path planner to quickly compute a closed and feasible path visiting all goals. Specifically, the model first predicts promising regions that potentially contain the optimal paths connecting two goals as a segmentation task. Simultaneously, estimations for pairwise distances between goals are conducted as a regression task by the neural networks, while the results construct a symmetric weight matrix for the TSP solver. Leveraging the TSP result, the path planner efficiently explores feasible paths guided by promising regions. We extensively evaluate the S&Reg model through simulations and compare it with the other sampling-based algorithms. The results demonstrate that our proposed model achieves superior performance in respect of computation time and solution cost, making it an effective solution for multi-goal path planning in obstacle environments. The proposed approach has the potential to be extended to other sampling-based algorithms for multi-goal path planning.
We propose a novel visual SLAM method that integrates text objects tightly by treating them as semantic features via fully exploring their geometric and semantic prior. The text object is modeled as a texture-rich planar patch whose semantic meaning is extracted and updated on the fly for better data association. With the full exploration of locally planar characteristics and semantic meaning of text objects, the SLAM system becomes more accurate and robust even under challenging conditions such as image blurring, large viewpoint changes, and significant illumination variations (day and night). We tested our method in various scenes with the ground truth data. The results show that integrating texture features leads to a more superior SLAM system that can match images across day and night. The reconstructed semantic 3D text map could be useful for navigation and scene understanding in robotic and mixed reality applications. Our project page: https://github.com/SJTU-ViSYS/TextSLAM .
Self-supervised monocular depth estimation has achieved impressive performance on outdoor datasets. Its performance however degrades notably in indoor environments because of the lack of textures. Without rich textures, the photometric consistency is too weak to train a good depth network. Inspired by the early works on indoor modeling, we leverage the structural regularities exhibited in indoor scenes, to train a better depth network. Specifically, we adopt two extra supervisory signals for self-supervised training: 1) the Manhattan normal constraint and 2) the co-planar constraint. The Manhattan normal constraint enforces the major surfaces (the floor, ceiling, and walls) to be aligned with dominant directions. The co-planar constraint states that the 3D points be well fitted by a plane if they are located within the same planar region. To generate the supervisory signals, we adopt two components to classify the major surface normal into dominant directions and detect the planar regions on the fly during training. As the predicted depth becomes more accurate after more training epochs, the supervisory signals also improve and in turn feedback to obtain a better depth model. Through extensive experiments on indoor benchmark datasets, the results show that our network outperforms the state-of-the-art methods. The source code is available at https://github.com/SJTU-ViSYS/StructDepth .
The task of classifying mammograms is very challenging because the lesion is usually small in the high resolution image. The current state-of-the-art approaches for medical image classification rely on using the de-facto method for ConvNets - fine-tuning. However, there are fundamental differences between natural images and medical images, which based on existing evidence from the literature, limits the overall performance gain when designed with algorithmic approaches. In this paper, we propose to go beyond fine-tuning by introducing a novel framework called MorphHR, in which we highlight a new transfer learning scheme. The idea behind the proposed framework is to integrate function-preserving transformations, for any continuous non-linear activation neurons, to internally regularise the network for improving mammograms classification. The proposed solution offers two major advantages over the existing techniques. Firstly and unlike fine-tuning, the proposed approach allows for modifying not only the last few layers but also several of the first ones on a deep ConvNet. By doing this, we can design the network front to be suitable for learning domain specific features. Secondly, the proposed scheme is scalable to hardware. Therefore, one can fit high resolution images on standard GPU memory. We show that by using high resolution images, one prevents losing relevant information. We demonstrate, through numerical and visual experiments, that the proposed approach yields to a significant improvement in the classification performance over state-of-the-art techniques, and is indeed on a par with radiology experts. Moreover and for generalisation purposes, we show the effectiveness of the proposed learning scheme on another large dataset, the ChestX-ray14, surpassing current state-of-the-art techniques.