Explore the vast range of titles available.

Drawing on the analytic of the “colonial matrix of power” developed by Aníbal Quijano within the Latin American modernity/coloniality research program, this article theorises how a system of coloniality underpins the structuring logic of artificial intelligence (AI) systems. We develop a framework for critiquing the regimes of global labour exploitation and knowledge extraction that are rendered invisible through discourses of the purported universality and objectivity of AI. ​​Through bringing the political economy literature on AI production into conversation with scholarly work on decolonial AI and the modernity/coloniality research program, we advance three main arguments. First, the global economic and political power imbalances in AI production are inextricably linked to the continuities of historical colonialism, constituting the colonial supply chain of AI. Second, this is produced through an international division of digital labour that extracts value from majority world labour for the benefit of Western technology companies. Third, this perpetuates hegemonic knowledge production through Western values and knowledge that marginalises non-Western alternatives within AI’s production and limits the possibilities for decolonising AI. By locating the production of AI systems within the colonial matrix of power, we contribute to critical and decolonial literature on the legacies of colonialism in AI and the hierarchies of power and extraction that shape the development of AI today.

This article provides a new typology for understanding human labour integrated into the production of artificial intelligence systems through data preparation and model evaluation. We call these forms of labour ‘AI data work’ and show how they are an important and necessary element of the artificial intelligence production process. We draw on fieldwork with an artificial intelligence data business process outsourcing centre specialising in computer vision data, alongside a decade of fieldwork with microwork platforms, business process outsourcing, and artificial intelligence companies to help dispel confusion around the multiple concepts and frames that encompass artificial intelligence data work including ‘ghost work’, ‘microwork’, ‘crowdwork’ and ‘cloudwork’. We argue that these different frames of reference obscure important differences between how this labour is organised in different contexts. The article provides a conceptual division between the different types of artificial intelligence data work institutions and the different stages of what we call the artificial intelligence data pipeline. This article thus contributes to our understanding of how the practices of workers become a valuable commodity integrated into global artificial intelligence production networks.

This study evaluates Melbourne's longstanding 'activity centres' (AC) policies-the first study to do so. It strongly suggests that, across the Melbourne metropolitan area, AC policies have had no effect on the propensity of people to work near their homes. The findings are robust to a number of validity hazards. The study does not warrant a wholesale abandonment of AC planning, but does signal that we may wish to question how we are currently going about transforming 'places' into 'centres'. For AC policies to be successful, designation as a 'centre' may be necessary, but is not sufficient.

This study evaluates Melbourne’s longstanding ‘activity centres’ (AC) policies—the first study to do so. It strongly suggests that, across the Melbourne metropolitan area, AC policies have had no effect on the propensity of people to work near their homes. The findings are robust to a number of validity hazards. The study does not warrant a wholesale abandonment of AC planning, but does signal that we may wish to question how we are currently going about transforming ‘places’ into ‘centres’. For AC policies to be successful, designation as a ‘centre’ may be necessary, but is not sufficient.

Multi-object tracking (MOT) is a critical technology in computer vision, designed to detect multiple targets in video sequences and assign each target a unique ID per frame. Existed MOT methods excel at accurately tracking multiple objects in real-time across various scenarios. However, these methods still face challenges such as poor noise resistance and frequent ID switches. In this research, we propose a novel ConsistencyTrack, joint detection and tracking(JDT) framework that formulates detection and association as a denoising diffusion process on perturbed bounding boxes. This progressive denoising strategy significantly improves the model's noise resistance. During the training phase, paired object boxes within two adjacent frames are diffused from ground-truth boxes to a random distribution, and then the model learns to detect and track by reversing this process. In inference, the model refines randomly generated boxes into detection and tracking results through minimal denoising steps. ConsistencyTrack also introduces an innovative target association strategy to address target occlusion. Experiments on the MOT17 and DanceTrack datasets demonstrate that ConsistencyTrack outperforms other compared methods, especially better than DiffusionTrack in inference speed and other performance metrics. Our code is available at https://github.com/Tankowa/ConsistencyTrack.

Significant advancements in video diffusion models have brought substantial progress to the field of text-to-video (T2V) synthesis. However, existing T2V synthesis model struggle to accurately generate complex motion dynamics, leading to a reduction in video realism. One possible solution is to collect massive data and train the model on it, but this would be extremely expensive. To alleviate this problem, in this paper, we reformulate the typical T2V generation process as a search-based generation pipeline. Instead of scaling up the model training, we employ existing videos as the motion prior database. Specifically, we divide T2V generation process into two steps: (i) For a given prompt input, we search existing text-video datasets to find videos with text labels that closely match the prompt motions. We propose a tailored search algorithm that emphasizes object motion features. (ii) Retrieved videos are processed and distilled into motion priors to fine-tune a pre-trained base T2V model, followed by generating desired videos using input prompt. By utilizing the priors gleaned from the searched videos, we enhance the realism of the generated videos' motion. All operations can be finished on a single NVIDIA RTX 4090 GPU. We validate our method against state-of-the-art T2V models across diverse prompt inputs. The code will be public.

The current text-to-video (T2V) generation has made significant progress in synthesizing realistic general videos, but it is still under-explored in identity-specific human video generation with customized ID images. The key challenge lies in maintaining high ID fidelity consistently while preserving the original motion dynamic and semantic following after the identity injection. Current video identity customization methods mainly rely on reconstructing given identity images on text-to-image models, which have a divergent distribution with the T2V model. This process introduces a tuning-inference gap, leading to dynamic and semantic degradation. To tackle this problem, we propose a novel framework, dubbed \\textbf{PersonalVideo}, that applies direct supervision on videos synthesized by the T2V model to bridge the gap. Specifically, we introduce a learnable Isolated Identity Adapter to customize the specific identity non-intrusively, which does not comprise the original T2V model's abilities (e.g., motion dynamic and semantic following). With the non-reconstructive identity loss, we further employ simulated prompt augmentation to reduce overfitting by supervising generated results in more semantic scenarios, gaining good robustness even with only a single reference image available. Extensive experiments demonstrate our method's superiority in delivering high identity faithfulness while preserving the inherent video generation qualities of the original T2V model, outshining prior approaches. Notably, our PersonalVideo seamlessly integrates with pre-trained SD components, such as ControlNet and style LoRA, requiring no extra tuning overhead.

While features of different scales are perceptually important to visual inputs, existing vision transformers do not yet take advantage of them explicitly. To this end, we first propose a cross-scale vision transformer, CrossFormer. It introduces a cross-scale embedding layer (CEL) and a long-short distance attention (LSDA). On the one hand, CEL blends each token with multiple patches of different scales, providing the self-attention module itself with cross-scale features. On the other hand, LSDA splits the self-attention module into a short-distance one and a long-distance counterpart, which not only reduces the computational burden but also keeps both small-scale and large-scale features in the tokens. Moreover, through experiments on CrossFormer, we observe another two issues that affect vision transformers' performance, i.e., the enlarging self-attention maps and amplitude explosion. Thus, we further propose a progressive group size (PGS) paradigm and an amplitude cooling layer (ACL) to alleviate the two issues, respectively. The CrossFormer incorporating with PGS and ACL is called CrossFormer++. Extensive experiments show that CrossFormer++ outperforms the other vision transformers on image classification, object detection, instance segmentation, and semantic segmentation tasks. The code will be available at: https://github.com/cheerss/CrossFormer.

Logit based knowledge distillation gets less attention in recent years since feature based methods perform better in most cases. Nevertheless, we find it still has untapped potential when we re-investigate the temperature, which is a crucial hyper-parameter to soften the logit outputs. For most of the previous works, it was set as a fixed value for the entire distillation procedure. However, as the logits from different samples are distributed quite variously, it is not feasible to soften all of them to an equal degree by just a single temperature, which may make the previous work transfer the knowledge of each sample inadequately. In this paper, we restudy the hyper-parameter temperature and figure out its incapability to distill the knowledge from each sample sufficiently when it is a single value. To address this issue, we propose Normalized Knowledge Distillation (NormKD), with the purpose of customizing the temperature for each sample according to the characteristic of the sample's logit distribution. Compared to the vanilla KD, NormKD barely has extra computation or storage cost but performs significantly better on CIRAR-100 and ImageNet for image classification. Furthermore, NormKD can be easily applied to the other logit based methods and achieve better performance which can be closer to or even better than the feature based method.

Recent self-training techniques have shown notable improvements in unsupervised domain adaptation for 3D object detection (3D UDA). These techniques typically select pseudo labels, i.e., 3D boxes, to supervise models for the target domain. However, this selection process inevitably introduces unreliable 3D boxes, in which 3D points cannot be definitively assigned as foreground or background. Previous techniques mitigate this by reweighting these boxes as pseudo labels, but these boxes can still poison the training process. To resolve this problem, in this paper, we propose a novel pseudo label refinery framework. Specifically, in the selection process, to improve the reliability of pseudo boxes, we propose a complementary augmentation strategy. This strategy involves either removing all points within an unreliable box or replacing it with a high-confidence box. Moreover, the point numbers of instances in high-beam datasets are considerably higher than those in low-beam datasets, also degrading the quality of pseudo labels during the training process. We alleviate this issue by generating additional proposals and aligning RoI features across different domains. Experimental results demonstrate that our method effectively enhances the quality of pseudo labels and consistently surpasses the state-of-the-art methods on six autonomous driving benchmarks. Code will be available at https://github.com/Zhanwei-Z/PERE.