Hasil untuk "physics.ins-det"

Yecheng Huang, Jiaxin Chen, Di Huang

This paper proposes a novel approach to object detection on drone imagery, namely Multi-Proxy Detection Network with Unified Foreground Packing (UFPMP-Det). To deal with the numerous instances of very small scales, different from the common solution that divides the high-resolution input image into quite a number of chips with low foreground ratios to perform detection on them each, the Unified Foreground Packing (UFP) module is designed, where the sub-regions given by a coarse detector are initially merged through clustering to suppress background and the resulting ones are subsequently packed into a mosaic for a single inference, thus significantly reducing overall time cost. Furthermore, to address the more serious confusion between inter-class similarities and intra-class variations of instances, which deteriorates detection performance but is rarely discussed, the Multi-Proxy Detection Network (MP-Det) is presented to model object distributions in a fine-grained manner by employing multiple proxy learning, and the proxies are enforced to be diverse by minimizing a Bag-of-Instance-Words (BoIW) guided optimal transport loss. By such means, UFPMP-Det largely promotes both the detection accuracy and efficiency. Extensive experiments are carried out on the widely used VisDrone and UAVDT datasets, and UFPMP-Det reports new state-of-the-art scores at a much higher speed, highlighting its advantages. The code is available at https://github.com/PuAnysh/UFPMP-Det.

Chenfeng Xu, Bichen Wu, Ji Hou et al.

We present NeRF-Det, a novel method for indoor 3D detection with posed RGB images as input. Unlike existing indoor 3D detection methods that struggle to model scene geometry, our method makes novel use of NeRF in an end-to-end manner to explicitly estimate 3D geometry, thereby improving 3D detection performance. Specifically, to avoid the significant extra latency associated with per-scene optimization of NeRF, we introduce sufficient geometry priors to enhance the generalizability of NeRF-MLP. Furthermore, we subtly connect the detection and NeRF branches through a shared MLP, enabling an efficient adaptation of NeRF to detection and yielding geometry-aware volumetric representations for 3D detection. Our method outperforms state-of-the-arts by 3.9 mAP and 3.1 mAP on the ScanNet and ARKITScenes benchmarks, respectively. We provide extensive analysis to shed light on how NeRF-Det works. As a result of our joint-training design, NeRF-Det is able to generalize well to unseen scenes for object detection, view synthesis, and depth estimation tasks without requiring per-scene optimization. Code is available at https://github.com/facebookresearch/NeRF-Det.

Sonny Ankjær Sahl

-

Guotao Xie, Zhiyuan Chen, Ming Gao et al.

Multi-modal fusion can take advantage of the LiDAR and camera to boost the robustness and performance of 3D object detection. However, there are still of great challenges to comprehensively exploit image information and perform accurate diverse feature interaction fusion. In this paper, we proposed a novel multi-modal framework, namely Point-Pixel Fusion for Multi-Modal 3D Object Detection (PPF-Det). The PPF-Det consists of three submodules, Multi Pixel Perception (MPP), Shared Combined Point Feature Encoder (SCPFE), and Point-Voxel-Wise Triple Attention Fusion (PVW-TAF) to address the above problems. Firstly, MPP can make full use of image semantic information to mitigate the problem of resolution mismatch between point cloud and image. In addition, we proposed SCPFE to preliminary extract point cloud features and point-pixel features simultaneously reducing time-consuming on 3D space. Lastly, we proposed a fine alignment fusion strategy PVW-TAF to generate multi-level voxel-fused features based on attention mechanism. Extensive experiments on KITTI benchmarks, conducted on September 24, 2023, demonstrate that our method shows excellent performance.

Liming Liu, Hongjian Zhang, Shengyang Zhou et al.

With the advent of the Internet of Things, self-powered wearable sensors have become increasingly prevalent in our daily lives. The utilization of piezoelectric composites to harness and sense surrounding mechanical vibrations has been extensively investigated during the last decades. However, the poor interface compatibility between ceramics nanofillers and polymers matrix, as well as low piezoelectric performance, still serves as a critical challenge. In this work, we employed Di(dioctylpyrophosphato) ethylene titanate (DET) as the coupling agent for modifying barium titanate (BTO) nanofillers. Compared to the BTO/PVDF counterpart, the DET-BTO/PVDF nanofibers exhibit an augmented content of piezoelectric β phase (~85.7%) and significantly enhanced stress transfer capability. The piezoelectric coefficient (d33) is up to ~40 pC/N, which is the highest value among reported BTO/PVDF composites. The piezoelectric energy harvesters (PEHs) present benign durability and attain a high instantaneous power density of 276.7 nW/cm2 at a matched load of 120 MΩ. Furthermore, the PEHs could sense various human activities, with the sensitivity as high as 0.817 V/N ranging from 0.05–0.1 N. This work proposes a new strategy to boosting the piezoelectric performance of PVDF-based composites via DET-doping ceramics nanoparticles, and in turn show significantly improved energy harvesting and sensing capability.

Zhigang Yang, Xiangyu Xia, Yiming Liu et al.

Convolutional neural networks (CNNs) and transformers have boosted the rapid growth of object detection in synthetic aperture radar (SAR) images. However, it is still a challenging task because SAR images usually have the characteristics of unclear contour, sidelobe interference, speckle noise, multiple scales, complex inshore background, etc. More effective feature extraction by the backbone and augmentation in the neck will bring a promising performance increment. In response, we make full use of the advantage of CNNs in extracting local features and the advantage of transformers in capturing long-range dependencies to propose a Swin Transformer-based detector for arbitrary-oriented SAR ship detection. Firstly, we incorporate a convolution-based local perception unit (CLPU) into the transformer structure to establish a powerful backbone. The local-perception-enhanced Swin Transformer (LP-Swin) backbone combines the local information perception ability of CNNs and the global feature extraction ability of transformers to enhance representation learning, which can extract object features more effectively and boost the detection performance. Then, we devise a cross-scale bidirectional feature pyramid network (CS-BiFPN) by strengthening the propagation and integration of both location and semantic information. It allows for more effective utilization of the feature extracted by the backbone and mitigates the problem of multi-scale ships. Moreover, we design a one-stage framework integrated with LP-Swin, CS-BiFPN, and the detection head of R3Det for arbitrary-oriented object detection, which can provide more precise locations for inclined objects and introduce less background information. On the SAR Ship Detection Dataset (SSDD), ablation studies are implemented to verify the effectiveness of each component, and competing experiments illustrate that our detector attains 93.31% in mean average precision (mAP), which is a comparable detection performance with other advanced detectors.

Md Humaion Kabir Mehedi, Nafisa Nawer, Shafi Ahmed et al.

Yang Cao, Yuanliang Jv, Dan Xu

Neural Radiance Fields (NeRF) have been adapted for indoor 3D Object Detection (3DOD), offering a promising approach to indoor 3DOD via view-synthesis representation. But its implicit nature limits representational capacity. Recently, 3D Gaussian Splatting (3DGS) has emerged as an explicit 3D representation that addresses the limitation. This work introduces 3DGS into indoor 3DOD for the first time, identifying two main challenges: (i) Ambiguous spatial distribution of Gaussian blobs -- 3DGS primarily relies on 2D pixel-level supervision, resulting in unclear 3D spatial distribution of Gaussian blobs and poor differentiation between objects and background, which hinders indoor 3DOD; (ii) Excessive background blobs -- 2D images typically include numerous background pixels, leading to densely reconstructed 3DGS with many noisy Gaussian blobs representing the background, negatively affecting detection. To tackle (i), we leverage the fact that 3DGS reconstruction is derived from 2D images, and propose an elegant solution by incorporating 2D Boundary Guidance to significantly enhance the spatial distribution of Gaussian blobs, resulting in clearer differentiation between objects and their background (please see fig:teaser). To address (ii), we propose a Box-Focused Sampling strategy using 2D boxes to generate object probability distribution in 3D space, allowing effective probabilistic sampling in 3D to retain more object blobs and reduce noisy background blobs. Benefiting from these innovations, 3DGS-DET significantly outperforms the state-of-the-art NeRF-based method, NeRF-Det++, achieving improvements of +6.0 on mAP@0.25 and +7.8 on mAP@0.5 for the ScanNet, and the +14.9 on mAP@0.25 for the ARKITScenes.

Tanja M. Gampfer, Victoria Schütz, Philip Schippers et al.

The continuous emergence of new psychoactive substances (NPS) attracted a great deal of attention within recent years. Lately, the two hallucinogenic NPS 1cP-LSD and 4-AcO-DET have appeared on the global market. Knowledge about their metabolism to identify potential metabolic targets for analysis and their cytotoxic properties is lacking. The aim of this work was thus to study their in vitro and in vivo metabolism in pooled human liver S9 fraction (pHLS9) and in zebrafish larvae (ZL) by means of liquid chromatography-high-resolution tandem mass spectrometry. Monooxygenases involved in the initial metabolic steps were elucidated using recombinant human isozymes. Investigations on their cytotoxicity were performed on the human hepatoma cell line HepG2 using a multiparametric, fluorescence-based high-content screening assay. This included measurement of CYP-enzyme mediated effects by means of the unspecific CYP inhibitor 1-aminbenzotriazole (ABT). Several phase I metabolites of both compounds and two phase II metabolites of 4-AcO-DET were produced in vitro and in vivo. After microinjection of 1cP-LSD into the caudal vein of ZL, three out of seven metabolites formed in pHLS9 were also detected in ZL. Twelve 4-AcO-DET metabolites were identified in ZL after exposure via immersion bath and five of them were found in pHLS9 incubations. Notably, unique metabolites of 4-AcO-DET were only produced by ZL, whereas 1cP-LSD specific metabolites were found both in ZL and in pHLS9. No toxic effects were observed for 1cP-LSD and 4-AcO-DET in HepG2 cells, however, two parameters were altered in incubations containing 4-AcO-DET together with ABT compared with incubations without ABT but in concentrations far above expected in vivo concentration. Further investigations should be done with other hepatic cell lines expressing higher levels of CYP enzymes.

Yanan Zhang, Jiaxin Chen, Di Huang

In autonomous driving, LiDAR point-clouds and RGB images are two major data modalities with complementary cues for 3D object detection. However, it is quite difficult to sufficiently use them, due to large inter-modal discrepancies. To address this issue, we propose a novel framework, namely Contrastively Augmented Transformer for multi-modal 3D object Detection (CAT-Det). Specifically, CAT-Det adopts a two-stream structure consisting of a Pointformer (PT) branch, an Imageformer (IT) branch along with a Cross-Modal Transformer (CMT) module. PT, IT and CMT jointly encode intra-modal and inter-modal long-range contexts for representing an object, thus fully exploring multi-modal information for detection. Furthermore, we propose an effective One-way Multimodal Data Augmentation (OMDA) approach via hierarchical contrastive learning at both the point and object levels, significantly improving the accuracy only by augmenting point-clouds, which is free from complex generation of paired samples of the two modalities. Extensive experiments on the KITTI benchmark show that CAT-Det achieves a new state-of-the-art, highlighting its effectiveness.

Chenxi Huang, Yuenan Hou, Weicai Ye et al.

NeRF-Det has achieved impressive performance in indoor multi-view 3D detection by innovatively utilizing NeRF to enhance representation learning. Despite its notable performance, we uncover three decisive shortcomings in its current design, including semantic ambiguity, inappropriate sampling, and insufficient utilization of depth supervision. To combat the aforementioned problems, we present three corresponding solutions: 1) Semantic Enhancement. We project the freely available 3D segmentation annotations onto the 2D plane and leverage the corresponding 2D semantic maps as the supervision signal, significantly enhancing the semantic awareness of multi-view detectors. 2) Perspective-aware Sampling. Instead of employing the uniform sampling strategy, we put forward the perspective-aware sampling policy that samples densely near the camera while sparsely in the distance, more effectively collecting the valuable geometric clues. 3)Ordinal Residual Depth Supervision. As opposed to directly regressing the depth values that are difficult to optimize, we divide the depth range of each scene into a fixed number of ordinal bins and reformulate the depth prediction as the combination of the classification of depth bins as well as the regression of the residual depth values, thereby benefiting the depth learning process. The resulting algorithm, NeRF-Det++, has exhibited appealing performance in the ScanNetV2 and ARKITScenes datasets. Notably, in ScanNetV2, NeRF-Det++ outperforms the competitive NeRF-Det by +1.9% in mAP@0.25 and +3.5% in mAP@0.50$. The code will be publicly at https://github.com/mrsempress/NeRF-Detplusplus.

Anna Lawaetz

The digital turn has influenced art practices and forms of expression, and has created among artists an increased awareness of the archive. The digital turn changes how archives are created, and new strategies for collecting need to be established to ensure that the cultural heritage is preserved. The study understands the performing arts archive as a living archive that is part of an artistic circuit. Through interviews with ten Danish-based performing artists who do not use the play text as a score for their work, the article examines what the artists themselves preserve and what kinds of record they need from the archive when re-staging or reinterpreting other artists’ work. In order to get an overview of the nature of the artists’ archives, two axes are used: one that measures the volume and one that measures the nature of the content. Through an investigation of what the performing artists preserve, particular emphasis is ascribed to documentation of the work, documentation of the process and documentation of the experience. In addition, a new axis that deals with the time-sensitivity of archives is proposed. It is not ideology that causes the performing arts not to be preserved by the individual artist in the study but a matter of chance, of economy in relation to recording and storage room. There is a tendency to save materials that make one’s own works appear as they were intended. Several of the artists have digital online archives, socalled ‘frontstage archives’, which differ from the digital archives they keep backstage, privately. A new trend can be seen among some of the artists, where the performance is documented by the audience at the performance, which calls into question the expressive potential of the material: is it part of the work itself or is it a perception of the work that can be equated with a review? In the case of the need for archival materials for re-staging or reinterpretation, there is not necessarily a need for the same things that the artists themselves have preserved: here the raw video documentation, work notes, the intention on the part of the artist in his/her own words, in some cases recordings from rehearsals and contact with the bodies that performed the work are requested, in order to gain insight into the method. It is very clear that for the interviewees there is a need for an interaction between certain types of archive fragments. Surprisingly enough, there was no correlation between artistic form of expression (durational, relational, wordless) and what was requested in relation to documentation. Finally, the study maps the artists’ thoughts on different reporting formats, where both economics and transparency about purpose come into play as important motivational factors.

L. Pedersen

In August 1947 Copenhagen hosted the first international congress of Egyptologists since the end of the Second World War. About thirty leading Egyptologists from the United States, Africa and Europe (with the exception of Germany) made their way to the Congress, which had important issues on the agenda, including the creation of an International Union of Egyptologists and the re-establishment of the most important international journals whose activities had ceased during the war. The atmosphere among the participants was good, but there was a fly in the ointment. The Danish host of the Congress, Professor of Egyptology C.E. Sander-Hansen, had failed to invite the head of the Glyptotek’s Egyptian department, Otto Koefoed-Petersen. Koefoed-Petersen was far from happy with that decision, and he therefore launched attacks against Sander- Hansen in several Danish newspapers, in which he suggested that Sander-Hansen and other Danish members of the host committee had had links with representatives of the German occupying power during the war. Where Koefoed-Petersen got this information from is uncertain, but the information was true. In August-September 1941 a meeting of Danish and German Egyptologists took place in Copenhagen. The main reason for the meeting was to address the challenges faced by the long-standing collaboration between the scientific academies in Berlin and Copenhagen regarding the publication of the Dictionary of the Egyptian Language, Wörterbuch der ägyptischen Sprache. The outbreak of war in September 1939 had made this work difficult, as the dictionary’s extensive amount of source texts (Zetteln) and archive in Berlin had been taken to safety, while several of the dictionary’s younger employees had been called up for military service. The meeting in Copenhagen was attended on the German side by the professor of Egyptology at the University of Berlin, Hermann Grapow, who came to Denmark on 29 August 1941 in the company of the director of the Prussian Academy of Sciences in Berlin, the Orientalist Helmuth Scheel and the Berlin-based Danish Egyptologist Wolja Erichsen. During the first days in Copenhagen, Grapow and Scheel met with, among others, the president of the German Scientific Institute in Copenhagen (the Deutsche Wissenschaftliche Institut), which had opened in May 1941, the Kiel professor Otto Scheel, and with representatives from the German embassy. On 1 September the University of Copenhagen’s Egyptological laboratory in the heart of Copenhagen hosted the first meeting between the Danish Egyptologists and Hermann Grapow. The Danish side was represented by the Nestor of Danish Egyptology, H.O. Lange, and the younger Egyptologists C.E. Sander-Hansen, Aksel Volten and Wolja Erichsen. Three topics were on the agenda: continued collaboration on the Egyptian dictionary in Berlin, C.E. Sander-Hansen’s future work on the late Berlin professor Kurt Sethe’s comments on the oldest known religious texts from Egypt – the Pyramid Texts – and the plan to publish a demotic dictionary. Two days later Grapow gave a lecture at the German Scientific Institute, where Sander-Hansen and Volten were among the many prominent members of the audience, which also included several representatives from the German embassy, led by the plenipotentiary Cecil von Renthe-Fink. H.O. Lange had originally agreed to participate but later changed his mind, citing poor health and challenges navigating safely in the dark as reasons for his cancellation. On 6 September C.E. Sander-Hansen, Erik Iversen and Wolja Erichsen met with Scheel and Grapow at the German Scientific Institute. The meeting, which had come about at the initiative of the Danes, had a more informative nature and revolved around Lange’s impending eightieth birthday in October 1943 and the opportunity to publish a Festschrift in his honour. Grapow and Scheel also had a number of other tasks in Copenhagen. In addition to several meetings with the various representatives of the German occupying power in Denmark, Grapow held, among other things, a meeting about another ongoing German project regarding the registration and inventory of German medieval manuscripts in Denmark with the head of the Prussian Academy’s Deutsche Kommission manuscript archive, Hans Werner Pyritz, who had come to Copenhagen on 2 September, and with a German lecturer at the University of Copenhagen, Günther Jungbluth. Pyritz also had the opportunity to give a well-attended lecture at the German Scientific Institute before the small German delegation left Denmark again on 7 September 1941. Several German government institutions in both Copenhagen and Berlin subsequently considered the Danish-German Egyptologists’ meeting in Copenhagen a success. However, it was not, as had been hoped from the German side, the starting point for a more in-depth collaboration between the German Scientific Institute and Danish intellectuals. After the Danish-German meeting in Copenhagen, difficulties continued for the Egyptian dictionary’s remaining employees in Berlin, Grapow and Erichsen. Because of the war, otherwise completed works could not be printed, and in 1943 conditions in Berlin had become so uncertain for Wolja Erichsen and his family that they left the German capital and settled in Denmark. Erichsen never returned to the dictionary work in Berlin. The plans to publish a Festschrift to H.O. Lange came to nothing when, after a short illness, Lange passed away in January 1943. The German lecturer Günther Jungbluth had hardly got much further with his work of inventorying the German medieval manuscripts at the Royal Library and the University Library when he was called up for military service in January 1942 and had to leave Denmark. The Danish-German gathering in Copenhagen in 1941 had no consequences for the participating Danish Egyptologists after the liberation in May 1945. This was primarily due to the fact that the Danish public never found out about it – or rather, only did so very late. In 1941 the Danish newspapers wrote neither about the meeting of the Danish and German Egyptologists nor about Grapow’s and Pyritz’s lectures at the German Scientific Institute, with a number of German and Danish notables among the audience. The Danish-German meeting was therefore forgotten until Koefoed-Petersen brought it up in connection with the public dispute with Sander-Hansen in the late summer of 1947. Otto Koefoed-Petersen undoubtedly found the visit of his Danish Egyptologist colleagues to the German Scientific Institute during the occupation inappropriate. By bringing the subject up in connection with the Egyptologists’ conference in 1947, he probably hoped to be able to bring the Danish participants, and not least C.E. Sander-Hansen, into disrepute. However, that did not happen. Many newspapers were critical of Sander-Hansen’s actions regarding Koefoed-Petersen’s lack of invitation to the Egyptology conference, but none of them was apparently prompted to investigate the otherwise precarious subject of the comings and goings of Sander-Hansen and his colleagues at the German Scientific Institute during the occupation.

Md. Mustak Un Nobi, M. Rifat, M. F. Mridha et al.

The guava plant is widely cultivated in various regions of the Sub-Continent and Asian countries, including Bangladesh, due to its adaptability to different soil conditions and climate environments. The fruit plays a crucial role in providing food security and nutrition for the human body. However, guava plants are susceptible to various infectious leaf diseases, leading to significant crop losses. To address this issue, several heavyweight deep learning models have been developed in precision agriculture. This research proposes a transfer learning-based model named GLD-Det, which is designed to be both lightweight and robust, enabling real-time detection of guava leaf disease using two benchmark datasets. GLD-Det is a modified version of MobileNet, featuring additional components with two pooling layers such as max and global average, three batch normalisation layers, three dropout layers, ReLU as an activation function with four dense layers, and SoftMax as a classification layer with the last lighter dense layer. The proposed GLD-Det model outperforms all existing models with impressive accuracy, precision, recall, and AUC score with values of 0.98, 0.98, 0.97, and 0.99 on one dataset, and with values of 0.97, 0.97, 0.96, and 0.99 for the other dataset, respectively. Furthermore, to enhance trust and transparency, the proposed model has been explained using the Grad-CAM technique, a class-discriminative localisation approach.

Guanglei Song, Jiahai Yang, Zhiliang Wang et al.

Fast IPv4 scanning significantly improves network measurement and security research. Nevertheless, it is infeasible to perform brute-force scanning of the IPv6 address space. Alternatively, one can find active IPv6 addresses through scanning the candidate addresses generated by state-of-the-art algorithms. However, the probing efficiency of such algorithms is often very low. In this paper, our objective is to improve the probing efficiency of IPv6 addresses. We first perform a longitudinal active measurement study and build a high-quality dataset, hitlist, including more than 1.95B IPv6 addresses distributed in 58.2K BGP prefixes and collected over 17 months period. Different from the previous works, we probe the announced BGP prefixes using a pattern-based algorithm. This results in a dataset without uneven address distribution and low active rates. Further, we propose an efficient address generation algorithm, DET, which builds a density space tree to learn high-density address regions of the seed addresses with linear time complexity and improves the active addresses’ probing efficiency. We then compare our algorithm DET against state-of-the-art algorithms on the public hitlist and our hitlist by scanning 50M addresses. Our analysis shows that DET increases the de-aliased active address ratio and active address (including aliased addresses) ratio by 10%, and 14%, respectively. Furthermore, we develop a fingerprint-based method to detect aliased prefixes. The proposed method for the first time directly verifies whether the prefix is aliased or not. Our method finds that 10.64% of the public aliased prefixes are false positive.

Yong Zhou, Silin Chen, Jiaqi Zhao et al.

Challenges still exist in the task of object detection in remote sensing images with densely distributed objects due to large variation in scale and neglect of the relative position and correlation. To address these issues, a correlation learning detector based on transformer (CLT-Det) is proposed for detecting dense objects in remote sensing images. A transformer attention module (TAM) is designed to improve the densely packed objects’ model representation ability by learning pixelwise attention with a transformer. To alleviate the semantic gap caused by the variations in scale, a feature refinement module (FRM) is proposed by improving the multiscale feature pyramid. A correlation transformer module (CTM) is proposed to extract correlation information and it encodes position information of dense objects’ features on the classification branch for fully using the position information and correlation among objects. Extensive experiments compared with several state-of-art methods on two challenging remote sensing datasets, namely, dataset for object detection in aerial images (DOTA) and HRSC2016, demonstrate that the proposed CLT-Det achieves promising and competitive performance.

Meng-Ting Chang, L. Tsai, Kyoko Nakagawa-Goto et al.

Acquired resistance to vemurafenib (PLX4032) is a thorny issue in BRAFV600E mutant melanoma therapy. Ferroptotic programmed cell death is a potential strategy for combating therapy-resistant cancers. This study uncovers the adaptation and abnormal upregulation of PUFAs and bioactive oxylipin metabolism in PLX4032 resistant melanoma cells. Phyto-sesquiterpene lactone, DET, and its derivative, DETD-35, induced lipid ROS accumulation and triggered ferroptotic cell death in PLX4032 sensitive (A375) and resistant (A375-R) BRAFV600E melanoma cells by reprogramming glutathione and primary metabolisms, lipid/oxylipin metabolism, and causing mitochondrial damage in which DETD-35 showed superior efficiency to DET. We discovered that DET and DETD-35 are a new type of GPX4 enzyme inhibitor through non-covalent binding. This study provides new insight into the therapeutic mechanisms of both DET and DETD-35 to combat PLX4032 sensitive/resistant BRAFV600E mutant melanomas via targeting GPX4 and ferroptosis.

A. Westphall

Sheng Xu, Yanjing Li, Bo-Wen Zeng et al.

Knowledge distillation (KD) has been proven to be useful for training compact object detection models. However, we observe that KD is often effective when the teacher model and student counterpart share similar proposal information. This explains why existing KD methods are less effective for 1-bit detectors, caused by a significant information discrepancy between the real-valued teacher and the 1-bit student. This paper presents an Information Discrepancy-aware strategy (IDa-Det) to distill 1-bit detectors that can effectively eliminate information discrepancies and significantly reduce the performance gap between a 1-bit detector and its real-valued counterpart. We formulate the distillation process as a bi-level optimization formulation. At the inner level, we select the representative proposals with maximum information discrepancy. We then introduce a novel entropy distillation loss to reduce the disparity based on the selected proposals. Extensive experiments demonstrate IDa-Det's superiority over state-of-the-art 1-bit detectors and KD methods on both PASCAL VOC and COCO datasets. IDa-Det achieves a 76.9% mAP for a 1-bit Faster-RCNN with ResNet-18 backbone. Our code is open-sourced on https://github.com/SteveTsui/IDa-Det.