ECCV 2024 Papers

"We present an inverse image-formation module that can enhance the robustness of existing visual SLAM pipelines for casually captured scenarios. Casual video captures often suffer from motion blur and varying appearances, which degrade the final quality of coherent 3D visual representation. We propose integrating the physical imaging into the SLAM system, which employs linear HDR radiance maps to collect measurements. Specifically, individual frames aggregate images of multiple poses along the camera trajectory to explain prevalent motion blur in hand-held videos. Additionally, we accommodate per-frame appearance variation by dedicating explicit variables for image formation steps, namely white balance, exposure time, and camera response function. Through joint optimization of additional variables, the SLAM pipeline produces high-quality images with more accurate trajectories. Extensive experiments demonstrate that our approach can be incorporated into recent visual SLAM pipelines using various scene representations, such as neural radiance fields or Gaussian splatting. Project website"

"With the development of Deep Neural Networks (DNNs), many efforts have been made to handle medical image segmentation. Traditional methods such as nnUNet train specific segmentation models on the individual datasets. Plenty of recent methods have been proposed to adapt the foundational Segment Anything Model (SAM) to medical image segmentation. However, they still focus on discrete representations to generate pixel-wise predictions, which are spatially inflexible and scale poorly to higher resolution. In contrast, implicit methods learn continuous representations for segmentation, which is crucial for medical image segmentation. In this paper, we propose I-MedSAM, which leverages the benefits of both continuous representations and SAM, to obtain better cross-domain ability and accurate boundary delineation. Since medical image segmentation needs to predict detailed segmentation boundaries, we designed a novel adapter to enhance the SAM features with high-frequency information during Parameter-Efficient Fine-Tuning (PEFT). To convert the SAM features and coordinates into continuous segmentation output, we utilize Implicit Neural Representation (INR) to learn an implicit segmentation decoder. We also propose an uncertainty-guided sampling strategy for efficient learning of INR. Extensive evaluations on 2D medical image segmentation tasks have shown that our proposed method with only 1.6M trainable parameters outperforms existing methods including discrete and implicit methods. The code will be available at: https://github.com/ucwxb/I-MedSAM."

"State-of-the-art scene flow methods broadly fail to describe the motion of small objects, and existing evaluation protocols hide this failure by averaging over many points. To address this limitation, we propose Bucket Normalized EPE, a new class-aware and speed-normalized evaluation protocol that better contextualizes error comparisons between object types that move at vastly different speeds. In addition, we propose TrackFlow, a frustratingly simple supervised scene flow baseline that combines a high-quality 3D object detector (trained using standard class re-balancing techniques) with a simple Kalman filter-based tracker. Notably, TrackFlow achieves state-of-the-art performance on existing metrics and shows large improvements over prior work on our proposed metric. Our results highlight that scene flow evaluation must be class and speed aware, and supervised scene flow methods must address point-level class imbalances. Our evaluation toolkit and code is available on GitHub."

"In this paper, we introduce a technique to estimate measured BRDFs from a sparse set of samples. Our approach offers accurate BRDF reconstructions that are generalizable to new materials. This opens the door to BRDF reconstructions from a variety of data sources. The success of our approach relies on the ability of hypernetworks to generate a robust representation of BRDFs and a set encoder that allows us to feed inputs of different sizes to the architecture. The set encoder and the hypernetwork also enable the compression of densely sampled BRDFs. We evaluate our technique both qualitatively and quantitatively on the well-known MERL dataset of 100 isotropic materials. Our approach accurately 1) estimates the BRDFs of unseen materials even for an extremely sparse sampling, 2) compresses the measured BRDFs into very small embeddings, e.g., 7D."

"Traditional deep learning models rely on methods such as softmax cross-entropy and ArcFace loss for tasks like classification and face recognition. These methods mainly explore angular features in a hyperspherical space, often resulting in entangled inter-class features due to dense angular data across many classes. In this paper, a new field of feature exploration is proposed known as HyperSpaceX which enhances class discrimination by exploring both angular and radial dimensions in multi-hyperspherical spaces, facilitated by a novel DistArc loss. The proposed DistArc loss encompasses three feature arrangement components: two angular and one radial, enforcing intra-class binding and inter-class separation in multi-radial arrangement, improving feature discriminability. Evaluation of HyperSpaceX framework for the novel representation utilizes a proposed predictive measure that accounts for both angular and radial elements, providing a more comprehensive assessment of model accuracy beyond standard metrics. Experiments across seven object classification and six face recognition datasets demonstrate state-of-the-art (SoTA) results obtained from HyperSpaceX, achieving up to a 20% performance improvement on large-scale object datasets in lower dimensions and up to 6% gain in higher dimensions."

"Recent advancements in text-to-image diffusion models have shown remarkable creative capabilities with textual prompts, but generating personalized instances based on specific subjects, known as subject-driven generation, remains challenging. To tackle this issue, we present a new hybrid framework called , which merges the benefits of optimization-based and direct-regression methods. operates in two stages: the Word Embedding Probe, which generates a robust initial word embedding using a fine-tuned encoder, and the Word Embedding Refinement, which further adapts the encoder to specific subject images by optimizing key parameters. This approach allows for effective and fast inversion of visual concepts into textual embedding, even from a single image, while maintaining the model’s generalization capabilities."

"Generating high-quality videos that synthesize desired realistic content is a challenging task due to their intricate high dimensionality and complexity. Several recent diffusion-based methods have shown comparable performance by compressing videos to a lower-dimensional latent space, using traditional video autoencoder architecture. However, such method that employ standard frame-wise 2D or 3D convolution fail to fully exploit the spatio-temporal nature of videos. To address this issue, we propose a novel hybrid video diffusion model, called HVDM, which can capture spatio-temporal dependencies more effectively. HVDM is trained by a hybrid video autoencoder which extracts a disentangled representation of the video including: (i) a global context information captured by a 2D projected latent, (ii) a local volume information captured by 3D convolutions with wavelet decomposition, and (iii) a frequency information for improving the video reconstruction. Based on this disentangled representation, our hybrid autoencoder provide a more comprehensive video latent enriching the generated videos with fine structures and details. Experiments on standard video generation benchmarks such as UCF101, SkyTimelapse, and TaiChi demonstrate that the proposed approach achieves state-of-the-art video generation quality, showing a wide range of video applications (e.g., long video generation, image-to-video, and video dynamics control). The source code and pre-trained models will be publicly available once the paper is accepted."

"Hyperspectral Imaging (HSI) plays an increasingly critical role in precise vision tasks within remote sensing, capturing a wide spectrum of visual data. Transformer architectures have significantly enhanced HSI task performance, while advancements in Transformer Architecture Search (TAS) have improved model discovery. To harness these advancements for HSI classification, we make the following contributions: i) We propose HyTAS, the first benchmark on transformer architecture search for Hyperspectral imaging, ii) We comprehensively evaluate 12 different methods to identify the optimal transformer over 5 different datasets, iii) We perform an extensive factor analysis on the Hyperspectral transformer search performance, greatly motivating future research in this direction. All benchmark materials are available at HyTAS."

"Text-to-image diffusion models have significantly advanced in conditional image generation. However, these models usually struggle with accurately rendering images featuring humans, resulting in distorted limbs and other anomalies. This issue primarily stems from the insufficient recognition and evaluation of limb qualities in diffusion models. To address this issue, we introduce AbHuman, the first large-scale synthesized human benchmark focusing on anatomical anomalies. This benchmark consists of 56K synthesized human images, each annotated with detailed, bounding-box level labels identifying 147K human anomalies in 18 different categories. Based on this, the recognition of human anomalies can be established, which in turn enhances image generation through traditional techniques such as negative prompting and guidance. To further boost the improvement, we propose HumanRefiner, a novel plug-and-play approach for the coarse-to-fine refinement of human anomalies in text-to-image generation. Specifically, HumanRefiner utilizes a self-diagnostic procedure to detect and correct issues related to both coarse-grained abnormal human poses and fine-grained anomaly levels, facilitating pose-reversible diffusion generation. Experimental results on the AbHuman benchmark demonstrate that HumanRefiner significantly reduces generative discrepancies, achieving a 2.9x improvement in limb quality compared to the state-of-the-art open-source generator SDXL and a 1.4x improvement over DALL-E 3 in human evaluations. Our AbHuman dataset is available at https://github.com/Enderfga/HumanRefiner."

"Computer vision-based re-identification (Re-ID) systems are increasingly being deployed for estimating population size in large image collections. However, the estimated size can be significantly inaccurate when the task is challenging or when deployed on data from new distributions. We propose a human-in-the-loop approach for estimating population size driven by a pairwise similarity derived from an off-the-shelf Re-ID system. Our approach, based on nested importance sampling, selects pairs of images for human vetting driven by the pairwise similarity, and produces asymptotically unbiased population size estimates with associated confidence intervals. We perform experiments on various animal Re-ID datasets and demonstrate that our method outperforms strong baselines and active clustering approaches. In many cases, we are able to reduce the error rates of the estimated size from around 80% using CV alone to less than 20% by vetting a fraction (often less than 0.002%) of the total pairs. The cost of vetting reduces with the increase in accuracy and provides a practical approach for population size estimation within a desired tolerance when deploying Re-ID systems.1 1 Code available at: https://github.com/cvl-umass/counting-clusters"

"Existing 2D-to-3D pose lifting networks suffer from poor performance in cross-dataset benchmarks. Although 2D keypoints joined by “stick-figure” limbs is the dominant trend, stick-figures do not preserve occlusion information that is inherent in an image, resulting in significant ambiguities that are ruled out when occlusion information is present. In addition, datasets with ground truth 3D poses are much harder to obtain in contrast to similar human annotated 2D datasets. To address these issues, we propose to replace stick figures with abstract images—figures with opaque limbs that preserve occlusion information while implicitly encoding joint locations. We then break down the pose estimation task into two stages: (1) Generating an abstract image from a real image, and (2) garnering the pose from the abstract image. Crucially, given the GT 3D keypoints for a particular pose, we can synthesize an arbitrary number of abstract images of the same pose as seen from arbitrary cameras, even without a part map. Given a set of 3D GT keypoints, this allows training of Stage (2) on an unlimited dataset without over-training, which in turn allows us to correctly interpret poses from arbitrary viewpoints not included in the original dataset. Additionally, our unlimited training of Stage 2 allows good generalizations across datasets, demonstrated through a significant improvement in cross-dataset benchmarks, while still showing competitive performance in same-dataset benchmark."

"Existing motion forecasting models, while making progress, struggle to bridge the gap between the source and target domains. Recent solutions often rely on an unrealistic assumption that the target domain remains stationary. Due to the ever-changing environment, however, the real-world test distribution may experience ongoing/continual shifts over time, leading to catastrophic forgetting and error accumulation when adapting to evolving domains. To solve these challenges, this work introduces HoCoTTA, a framework for homeostatic continual test-time adaptation. It aligns with the knowledge distillation and parameter isolation paradigm, enabling the identification of domain-invariant and domain-specific knowledge, where the former is shared (to be retained) in continual TTA across domains, while the latter needs to be updated. Specifically, we propose a multi-domain homeostasis assessment to estimate the uncertainty of the current model parameter when faced with novel-domain samples. Then, the Fisher information matrix is computed to measure the parameter sensitivity, with larger indicating the domain-sensitive parameter, and vice versa. Moreover, we propose an isolated parameter optimization strategy to update those domain-specific parameters to adapt to the new-domain, while preserving the invariant ones. In our experimental result, HoCoTTA outperforms the state-of-the-art approaches on several benchmarks, especially excelling in addressing continuous domain drifts, achieving a large improvement."

"We introduce a new hair modeling method that uses a dual representation of classical hair strands and 3D Gaussians to produce accurate and realistic strand-based reconstructions from multi-view data. In contrast to recent approaches that leverage unstructured Gaussians to model human avatars, our method reconstructs the hair using 3D polylines, or strands. This fundamental difference allows the use of the resulting hairstyles out-of-the-box in modern computer graphics engines for editing, rendering, and simulation. Our 3D lifting method relies on unstructured Gaussians to generate multi-view ground truth data to supervise the fitting of hair strands. The hairstyle itself is represented in the form of the so-called strand-aligned 3D Gaussians. This representation allows us to combine strand-based hair priors, which are essential for realistic modeling of the inner structure of hairstyles, with the differentiable rendering capabilities of 3D Gaussian Splatting. Our method, named Gaussian Haircut, is evaluated on synthetic and real scenes and demonstrates state-of-the-art performance in the task of strand-based hair reconstruction. For more results, please refer to our project page: https://eth-ait.github.io/GaussianHaircut."

"Instructional videos are a common source for learning text-video or even multimodal representations by leveraging subtitles extracted with automatic speech recognition systems (ASR) from the audio signal in the videos. However, in contrast to human-annotated captions, both speech and subtitles naturally differ from the visual content of the videos and thus provide only noisy supervision. As a result, large-scale annotation-free web video training data remains sub-optimal for training text-video models. In this work, we propose to leverage the capabilities of large language models (LLMs) to obtain high-quality video descriptions aligned with videos at scale. Specifically, we prompt an LLM to create plausible video captions based on ASR subtitles of instructional videos. To this end, we introduce a prompting method that is able to take into account a longer text of subtitles, allowing us to capture the contextual information beyond one single sentence. We further prompt the LLM to generate timestamps for each produced caption based on the timestamps of the subtitles and finally align the generated captions to the video temporally. In this way, we obtain human-style video captions at scale without human supervision. We apply our method to the subtitles of the HowTo100M dataset, creating a new large-scale dataset, HowToCaption. Our evaluation shows that the resulting captions not only significantly improve the performance over many different benchmark datasets for zero-shot text-video retrieval and video captioning, but also lead to a disentangling of textual narration from the audio, boosting the performance in text-video-audio tasks.1 1 All data and code is available at https://github.com/ninatu/howtocaption."

"Recently, it was shown that the role of the teacher in knowledge distillation (KD) is to provide the student with an estimate of the true Bayes conditional probability density (BCPD). Notably, the new findings propose that the student’s error rate can be upper-bounded by the mean squared error (MSE) between the teacher’s output and BCPD. Consequently, to enhance KD efficacy, the teacher should be trained such that its output is close to BCPD in MSE sense. This paper elucidates that training the teacher model with MSE loss equates to minimizing the MSE between its output and BCPD, aligning with its core responsibility of providing the student with a BCPD estimate closely resembling it in MSE terms. In this respect, through a comprehensive set of experiments, we demonstrate that substituting the conventional teacher trained with cross-entropy loss with one trained using MSE loss in state-of-the-art KD methods consistently boosts the student’s accuracy, resulting in improvements of up to 2.6%. The code for this paper is publicly available at: https://github.com/ECCV2024MSE/ ECCV_MSE_Teacher."

"Do our facial expressions change when we speak over video calls? Given two unpaired sets of videos of people, we seek to automatically find spatio-temporal patterns that are distinctive of each set. Existing methods use discriminative approaches and perform post-hoc explainability analysis. Such methods are insufficient as they are unable to provide insights beyond obvious dataset biases, and the explanations are useful only if humans themselves are good at the task. Instead, we tackle the problem through the lens of generative domain translation: our method generates a detailed report of learned, input-dependent spatio-temporal features and the extent to which they vary between the domains. We demonstrate that our method can discover behavioral differences between conversing face-to-face (F2F) and on video-calls (VCs). We also show the applicability of our method on discovering differences in presidential communication styles. Additionally, we are able to predict temporal change-points in videos that decouple expressions in an unsupervised way, and increase the interpretability and usefulness of our model. Finally, our method, being generative, can be used to transform a video call to appear as if it were recorded in a F2F setting. Experiments and visualizations show our approach is able to discover a range of behaviors, taking a step towards deeper understanding of human behaviors. Video results, code and data can be found at facet.cs.columbia.edu."

"This work focuses on benchmarking the capabilities of vision large language models (VLLMs) in visual reasoning. Different from prior studies, we shift our focus from evaluating standard performance to introducing a comprehensive safety evaluation suite Unicorn, covering out-of-distribution (OOD) generalization and adversarial robustness. For the OOD evaluation, we present two novel visual question-answering (VQA) datasets, each with one variant, designed to test model performance under challenging conditions. In exploring adversarial robustness, we propose a straightforward attack strategy for misleading VLLMs to produce visual-unrelated responses. Moreover, we assess the efficacy of two jailbreaking strategies, targeting either the vision or language input of VLLMs. Our evaluation of 22 diverse models, ranging from open-source VLLMs to GPT-4V and Gemini Pro, yields interesting observations: 1) Current VLLMs struggle with OOD texts but not images, unless the visual information is limited; and 2) These VLLMs can be easily misled by deceiving vision encoders only, and their vision-language training often compromise safety protocols. We release this safety evaluation suite at https://github.com/UCSC-VLAA/vllm-safety-benchmark."

"A de facto standard approach in solving computer vision tasks is to use a common high-resolution camera and choose its placement on an agent based on human intuition. On the other hand, extremely simple and well-designed visual sensors found throughout nature allow many organisms to exhibit diverse, complex behaviors. In this work, motivated by these examples, we raise the following questions: 1. How effective simple visual sensors are in solving vision tasks? 2. What role does their design play in their effectiveness? We explore simple sensors with resolutions as low as one-by-one pixel, representing a single photoreceptor. First, we demonstrate that just a few photoreceptors can be enough to solve different vision tasks, such as visual navigation and continuous control, reasonably well, i.e., with the performance significantly better than that of a blind agent and comparable to a high-resolution camera. Second, we show that the design of these simple visual sensors plays a crucial role in their ability to provide useful information and successfully solve these tasks. To find a well-performing design, we present a computational design optimization algorithm and evaluate its effectiveness across different tasks and domains, showing promising results. Finally, we conduct a human survey to evaluate the effectiveness of intuitive designs engineered manually by humans, showing that the computationally found design is among the best designs in most cases."

"Typical projectors are designed to programmably display 2D content at a single depth. In this work, we explore how to engineer a depth-varying projector system that is capable of forming desired patterns at multiple depths. To this end, we leverage a holographic approach, but a naı̈ve implementation of such a system is limited in its depth programmability. Inspired by recent work in near-eye displays, we add a lens array to a holographic projector to maximize the depth variation of the projected content, for which we propose an optimization-driven calibration method. We demonstrate a number of applications using this system, including novel 3D interfaces for future wearables, privacy-preserving projection, depth sensing, and light curtains."