question_id stringlengths 2 6 | answer stringlengths 0 1.31k | images listlengths 1 1 | problem stringlengths 258 863 |
|---|---|---|---|
202062 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/209e9e9c3a2e4a399c44b6aa8be659d6_ca0aeae4acfb43358c2af8fe637f2aa3.jpg"
] | Would <c3,CAM_FRONT_LEFT,[151, 278]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
11816 | <c3,CAM_BACK,[468, 866]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/0c13be31e9854d97960d1af570db433f_186ec6735ca942e3b660c56ec9d1861b.jpg"
] | Except for the ego vehicle, what object would consider <c1,CAM_FRONT,[451, 291]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
347405 | No, there are no traffic elements in the front view. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/93032908bafd484ea10a39b0d5d3fff6_36be1ff670e84058a1d63d419f7909a1.jpg"
] | Is there any traffic element in the front view? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
104273 | <c2,CAM_BACK,[553, 764]> is at the back of <c1,CAM_FRONT,[489, 262]>, <c3,CAM_FRONT,[499, 263]> is at the back of <c1,CAM_FRONT,[489, 262]>, <c4,CAM_FRONT,[592, 262]> is at the back of <c1,CAM_FRONT,[489, 262]>, <c5,CAM_FRONT,[522, 263]> is at the back of <c1,CAM_FRONT,[489, 262]>, <c3,CAM_FRONT,[499, 263]> is at the front of <c2,CAM_BACK,[553, 764]>, <c4,CAM_FRONT,[592, 262]> is at the front of <c2,CAM_BACK,[553, 764]>, <c5,CAM_FRONT,[522, 263]> is at the front of <c2,CAM_BACK,[553, 764]>, <c4,CAM_FRONT,[592, 262]> is at the back of <c3,CAM_FRONT,[499, 263]>, <c5,CAM_FRONT,[522, 263]> is at the back of <c3,CAM_FRONT,[499, 263]> and <c5,CAM_FRONT,[522, 263]> is at the front of <c4,CAM_FRONT,[592, 262]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/3289046b886f4c98a676bd7e6a3c6ef1_7886bd0692c742b292950bda16778d35.jpg"
] | What is the relative positioning of the important objects in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
145588 | Keep going at the same speed. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/d4c560022dda448dabb0b60fb62dd6ba_f35b47134f3f40b4a257de070dbb3ed1.jpg"
] | Based on <c5,CAM_FRONT,[476, 215]> in this scene, what is the most possible action of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
180686 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/d654cae175de44319ace178994ed0329_b75527be10fd4c1da8cafc73f51bae09.jpg"
] | Will <c4,CAM_BACK,[523, 852]> change its motion state based on <c1,CAM_FRONT,[623, 349]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
308264 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/57f5503bf63144b8bf450315e860f58e_792c4d8df88f4bbbae76bbc2a2ad6d79.jpg"
] | Is <c2,CAM_FRONT,[445, 335]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
195298 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/7bcc4753bdb64d2cb407b55cc4e4527f_34d47c84b8bf4deba19b63fd38e0a907.jpg"
] | Which object is most likely to be occluded by <c2,CAM_FRONT_RIGHT,[935, 312]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
221607 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/edf6a55a70114f6296143ed471251804_6ccd4899d7504b0d8a155c111fd8b3cc.jpg"
] | Which object is most likely to be occluded by <c2,CAM_FRONT,[544, 271]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
123141 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/fb255bba9b6a4f37a8894e336d2136f1_f062b40008ad4c67bd781d175c5dbcb1.jpg"
] | Is <c1,CAM_BACK,[466, 807]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
286366 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/52b30ecd104a4f4eab6f8f5684a73e56_7b061b925bf94b7bb630dfde82301135.jpg"
] | Are there moving cars to the front of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
77714 | Yes, there are traffic elements in the front view. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/76b36918bd4c487581a4196ffd1b1434_ebd0d5af0b834e37818e62b11a7eafcd.jpg"
] | Is there any traffic element in the front view? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
98784 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/727214583a2a478094d4f093d20c7060_bbe374bd1f494aef834da6fee932d004.jpg"
] | Would <c6,CAM_FRONT_LEFT,[157, 370]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
22340 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/f36162eac28641a0978032ff89e0f752_c53408ed2e4c4f9c84903a7d1944f6c5.jpg"
] | Is <c4,CAM_BACK_RIGHT,[954, 814]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
205498 | One of the pedestrians is moving and two are standing. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c24317a5c0cb4f5c9fd31740eb1152f2_b96247bf95f3404c805bde586478e510.jpg"
] | What is the status of the pedestrians that are to the back left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
84268 | The action is to none, the reason is there is no safety issue. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/f612aec0c3e34e32b2e9e48670cdecca_9075855f390344c98fcdcdc471b93251.jpg"
] | Based on the observations of <c3,CAM_BACK,[521, 809]>, what are possible actions to be taken by <c2,CAM_BACK,[372, 848]>? What is the reason? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
106851 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e631037169574c67944176bf079ee75c_42cba5110540498cabdd2bcd1bacf4f3.jpg"
] | Is <c5,CAM_FRONT,[551, 279]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
177308 | Low. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/019b370afb444a2e8ce85654898307c8_be7cbdbdf0bc47ecbe32cc758796bcd6.jpg"
] | What is the probability of colliding with <c2,CAM_FRONT_LEFT,[170, 347]> after the ego vehicle decelerates and goes straight? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
272161 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/448cf480b0b8400a86881d28c2c5f734_a3e6096501ba403f8bb9b3b26c1db065.jpg"
] | Is <c2,CAM_BACK,[567, 750]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
31712 | There is one traffic cone to the back right of the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/15e1fa06e30e438a98430cc1fd0e8a69_d64f9d23a6ac4d808dde46bfaf7069d5.jpg"
] | What are objects to the back right of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
195053 | A | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/7bcc4753bdb64d2cb407b55cc4e4527f_e9b6aff35b6345618912329bd1ec142d.jpg"
] | What is the moving status of object <c5,CAM_FRONT,[456, 303]>? Please select the correct answer from the following options: A. Going ahead. B. Turn left. C. Backward. D. Back up. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
253226 | Stationary. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/6ab3d1e9476d4cd89d4949d69d056901_2000ce82d2bd4f6c93ec5948736897a9.jpg"
] | What is the observed status of object <c1,CAM_FRONT_LEFT,[106, 290]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
350057 | Red light. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/f55c8c62235146d895f611176fd0a5dc_1cffb2ff65ac4ee7acdc4c23bcf9baf8.jpg"
] | What is the visual description of <c4,CAM_FRONT,[533, 140]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
35852 | None. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/53e8446852bf488bb1b09ae032918bbd_b3e80341269c443693a727b40c6efc63.jpg"
] | Except for the ego vehicle, what object would consider <c1,CAM_BACK_RIGHT,[842, 751]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
313723 | Unknown. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e233467e827140efa4b42d2b4c435855_93f4f8ac1145418cadcf14dfd2a9cf78.jpg"
] | What is the target action of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
120710 | One bicycle is without a rider. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c0990deaa8954a54afc86f7710f349a1_188ae235d3e14705b8b6323b8c90b67d.jpg"
] | What is the status of the bicycle that is to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
163170 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/7f8714d8ca814914bf4ff3ed30123db5_63e38d44915b441498de5af02f452614.jpg"
] | Would <c6,CAM_BACK,[493, 758]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
221927 | Back up. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/edf6a55a70114f6296143ed471251804_598e7958803149318bed3c7c6b288833.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c3,CAM_BACK,[498, 799]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
7274 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/6207269cb9474c14b01f9610b2ee30aa_baa013dbb58c4e4d8e2d766353decfa8.jpg"
] | Which object is most likely to be occluded by <c5,CAM_BACK,[388, 765]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
325914 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/37a37b06ba5e4f978fc6cb77e90f2498_f769bbc6bcd74c69af94fba05d0da4b2.jpg"
] | Would <c1,CAM_BACK,[414, 826]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
16273 | The ego vehicle. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e1e664292aa144bc8d0d5d6441df084c_aebee391061f40b2ac06ca4bc9db8087.jpg"
] | What object would consider <c6,CAM_FRONT,[457, 128]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
111704 | Both are traffic signs. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/9047b53fd41540649dce014a128cbe1b_0f8910c8005d43029efa7de74e59ab44.jpg"
] | Are <c2,CAM_FRONT_LEFT,[268, 247]> and <c3,CAM_FRONT,[432, 331]> traffic signs? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
195196 | Firstly, notice that <c4,CAM_FRONT_LEFT,[164, 275]>. The object is stationary, so the ego vehicle should continue ahead at the same speed. Secondly, notice that <c1,CAM_FRONT,[565, 342]>. The object is a traffic sign, so the ego vehicle should continue ahead at the same speed. Thirdly, notice that <c2,CAM_FRONT,[437, 257]>. The object is stationary, so the ego vehicle should continue ahead at the same speed. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/7bcc4753bdb64d2cb407b55cc4e4527f_2e200ef0c5b24a4c981197fcb5a8fdb5.jpg"
] | What object should the ego vehicle notice first when the ego vehicle is getting to the next possible location? What is the state of the object that is first noticed by the ego vehicle and what action should the ego vehicle take? What object should the ego vehicle notice second when the ego vehicle is getting to the next possible location? What is the state of the object perceived by the ego vehicle as second and what action should the ego vehicle take? What object should the ego vehicle notice third? What is the state of the object perceived by the ego vehicle as third and what action should the ego vehicle take? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
348377 | There is a person wearing a white shirt to the back right of the ego vehicle, a blue and white bus to the front of the ego vehicle, a blue and white bus to the front of the ego vehicle, a person wearing black pants to the front right of the ego vehicle, and a black sedan to the back of the ego vehicle. The IDs of these objects are <c1,CAM_BACK_RIGHT,[841, 787]>, <c2,CAM_FRONT,[652, 219]>, <c3,CAM_FRONT,[540, 254]>, <c4,CAM_FRONT_RIGHT,[712, 299]>, and <c5,CAM_BACK,[507, 773]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/14b96c255dee4a838a8d401302ac2852_b4849b201f024ca5bf68ab298b379ce7.jpg"
] | What are the important objects in the current scene? Those objects will be considered for the future reasoning and driving decision. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
176878 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/8dd4349e80c345afac455cf76074f028_31f05f62c19445328b2a0b91178a9fc6.jpg"
] | Is <c2,CAM_FRONT,[481, 268]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
203989 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/82aef599650d462db73731b7ff40918b_071f9d7f2c4f48fcb72fc17756bedfa7.jpg"
] | Is <c3,CAM_BACK,[452, 801]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
290182 | None. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/dbe2eceb3e4649289366f674c2ba4274_74377123547c4b56a22d0a345f22578b.jpg"
] | What object would consider <c1,CAM_BACK,[386, 890]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
64446 | Stationary. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/1623e7fab5174f2a84cf1b539233e3a6_021fe3dfd6554cffb439538942767ac7.jpg"
] | What is the future state of <c1,CAM_FRONT,[419, 295]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
330899 | The action is to keep going at the same speed. The reason is that there is no safety issue, and the probability is high. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/a5d7cc4bf3404cccb89932d94ccb9da6_24c6e0108b5844778df82c055c62cece.jpg"
] | What actions could the ego vehicle take based on <c4,CAM_FRONT_LEFT,[255, 286]>? Why take this action and what's the probability? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
265647 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/ced236f5da704cbcb41249bb6ecc8b68_8039886af4784ae6ae8c17c6ccdda218.jpg"
] | Will <c3,CAM_BACK,[470, 844]> change its motion state based on <c1,CAM_FRONT_RIGHT,[849, 325]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
41848 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/4a4d09496d594c7385a1ec227bdb3a7f_815e2b1b6309483bb1281dc6f4a74aab.jpg"
] | Are there parked cars to the back left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
16440 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e1e664292aa144bc8d0d5d6441df084c_e44bf485e8854fceacc6a052a1a60ea4.jpg"
] | Are there traffic cones to the back left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
127731 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/dc9ef63b61394d3d8effeb51919f2144_7d04e12148d84552b8dd7a7ba79d1b29.jpg"
] | Which object is most likely to be occluded by <c1,CAM_BACK_LEFT,[68, 729]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
311362 | There are many cars and one truck behind the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c7604a179ea94e45857ec2c16416e5ad_6250acbfaa914921a7f834eabc93f42c.jpg"
] | What are objects to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
51199 | The ego vehicle. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/4bcb814456f044919bc052c208dbebc5_df35f62944f84401b3ca3ab9ad58e0ac.jpg"
] | What object would consider <c5,CAM_FRONT_RIGHT,[825, 332]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
219855 | Three cars are parked. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/5557f4edc1464442812293e2ea90a586_3c313f517d6b4d5bafb8929ddd7b46fe.jpg"
] | What is the status of the cars that are to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
311172 | There are many cars, two trucks, one construction vehicle, and one pedestrian in front of the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c7604a179ea94e45857ec2c16416e5ad_6ef59ea4f03c43828715154592f647ff.jpg"
] | What are objects to the front of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
75595 | There is one moving car in front of the ego car, one moving pedestrian to the front left of the ego car, and two moving pedestrians behind the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/4deb245e67374cd88e64d6557b652841_fa5606858493496e96f85a1535bc283d.jpg"
] | Please describe the current scene. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
328961 | There are many traffic elements in the front view. The information of these traffic elements is [(<ref>traffic light</ref><box>[[512, 231, 513, 241]]</box>,unknown), (<ref>traffic light</ref><box>[[520, 232, 521, 241]]</box>,unknown), (<ref>traffic light</ref><box>[[535, 237, 538, 245]]</box>,unknown), (<ref>traffic light</ref><box>[[543, 239, 544, 246]]</box>,unknown), (<ref>traffic light</ref><box>[[541, 250, 542, 254]]</box>,unknown), (<ref>traffic light</ref><box>[[545, 249, 546, 254]]</box>,unknown), (<ref>traffic light</ref><box>[[549, 249, 550, 254]]</box>,green), (<ref>traffic light</ref><box>[[558, 255, 559, 259]]</box>,unknown)]. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/5560a973257e407b8d5bf9fb92b1e0f3_782ec2ef376b4443951bd662fd664296.jpg"
] | Identify all the traffic elements in the front view, categorize them, determine their status, and predict the bounding box around each one. The output should be a list formatted as (c,s), where c represents the object with bounding box, s denotes the status. |
352380 | The action is to turn left, the reason is there is no safety issue. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c04cdbbe744c4da08d3df61c47d5cc94_06d948570939439182e426e8791ef096.jpg"
] | Based on the observations of <c2,CAM_BACK,[494, 821]>, what are possible actions to be taken by <c3,CAM_FRONT_RIGHT,[794, 282]>? What is the reason? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
235931 | Sharp right turn. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/5a1d2867bb504c888a628ad549cc10b6_2af816b44c784a8883f0aa82ea2261c9.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c1,CAM_BACK,[429, 833]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
205281 | Go straight. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/4772f5283211444eaabd463cd341d360_acb2820ed7264fe793bbd800a66051b9.jpg"
] | What is the target action of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
85582 | Moderate right turn. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c31adab2faea4db6ae5673eb8a507d00_ec312eac9ffa401f91dd63c89a3040c4.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c3,CAM_FRONT_RIGHT,[673, 230]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
141724 | The ego vehicle. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/4c5c675340d14cd88a65b5e22eb9c9c1_f66de92b1e8a4ed280243c816bcced43.jpg"
] | What object would consider <c5,CAM_FRONT_LEFT,[221, 344]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
179058 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e02aee6f1fec4d13ad77044466da5fb4_5be27477c4884ed4a49dda31bb4e0612.jpg"
] | Which object is most likely to be occluded by <c4,CAM_FRONT,[427, 272]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
95271 | <c3,CAM_FRONT,[565, 313]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/d6e6f5d622474271b695f4c7f07df7bb_277a2473fcd94768b180b511dc0b601b.jpg"
] | Except for the ego vehicle, what object would consider <c5,CAM_FRONT,[547, 298]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
280472 | Brake suddenly. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/75a4ec12042542149b0a77a0a10d6330_d4b313b4957044559e6618ecc147474e.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c4,CAM_BACK,[448, 746]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
250164 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/782f9486adbd4546a492e69b6697ef70_59a32e24ef6c413bb81af6cfb9c7d608.jpg"
] | Which object is most likely to be occluded by <c1,CAM_BACK,[550, 791]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
236879 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/798e8504b4364d378270333a349ef508_f31932ce754b408db390251167423dcc.jpg"
] | Which object is most likely to be occluded by <c2,CAM_FRONT,[482, 272]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
140500 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/277af8c7487f4f24bd2953d4dc8d11ff_e2aa12c975564954bc6e08250349369c.jpg"
] | Is <c2,CAM_FRONT,[525, 269]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
29788 | D | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/fee7ac5f5c1e4943877cd20d6e069427_60ae5aad4543441ea39567642f2b4dfb.jpg"
] | What is the moving status of object <c1,CAM_FRONT,[479, 258]>? Please select the correct answer from the following options: A. Bending over. B. Backward. C. Stopped. D. Going ahead. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
64510 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/1623e7fab5174f2a84cf1b539233e3a6_a385bb47325647bd804731c389a96097.jpg"
] | Are there parked cars to the front right of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
66110 | A man wearing a white shirt is the visual description of <c4,CAM_BACK,[512, 787]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/9a81caa3d5134d7f87ee4786ccef68f7_362982ee86f64d3cb1e68071c7004e20.jpg"
] | What is the visual description of <c4,CAM_BACK,[512, 787]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
66939 | Stationary. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/f634de95cc7043b8b38ceaac67d472cf_8bdad602f01d4e1eae328966bbbb91e0.jpg"
] | What is the observed status of object <c3,CAM_BACK,[511, 798]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
275220 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/bed12977df5a41dcb97ac69eeae3682c_92f79f1bd8274ff69cc2c15a0aeeab68.jpg"
] | Are there parked construction vehicles to the back right of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
183080 | Stationary. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/97349fd50e694a5f994b0954047b96d0_a80a4a8b304a494081850d57f71d9804.jpg"
] | What is the observed status of object <c6,CAM_FRONT,[515, 275]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
230098 | Silver SUV. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/64e1ad8a976542ab86c427d714550d06_0490e347db7447fb834f1c3fb282db5d.jpg"
] | What is the visual description of <c3,CAM_FRONT_LEFT,[231, 274]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
40790 | Moving. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c0cb50847c3949a0b4baf2e200ac897c_68316c596bf8486589c60a6663bfb55d.jpg"
] | What is the observed status of object <c3,CAM_FRONT_LEFT,[144, 302]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
312399 | Black SUV. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/b57e1a48e584451f8f966ab083c8d07d_00d479f92c074090bfb4169d64464ee1.jpg"
] | What is the visual description of <c1,CAM_FRONT,[438, 283]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
182551 | There is one bicycle, one car, and many pedestrians to the front left of the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/1ad821fcfc9f4c24bb879ec7d5ba0ec1_135227f8855840da889dac07f55345f4.jpg"
] | What are objects to the front left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
365887 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/d4c5db6faa9f443795ad86561e9e556e_0d7c9d2d8557426dbc87ce0b21f7b5d1.jpg"
] | Would <c1,CAM_BACK_LEFT,[249, 757]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
325845 | The ego vehicle. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/37a37b06ba5e4f978fc6cb77e90f2498_caf95326f8cf44acacb39abf00c05364.jpg"
] | What object would consider <c2,CAM_FRONT,[549, 269]> to be most relevant to its decision? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
165800 | <c2,CAM_FRONT_LEFT,[159, 400]> is behind <c1,CAM_FRONT,[533, 251]>, <c3,CAM_FRONT,[562, 273]> is behind <c1,CAM_FRONT,[533, 251]>, <c4,CAM_FRONT,[362, 330]> is behind <c1,CAM_FRONT,[533, 251]>, <c5,CAM_FRONT_RIGHT,[909, 393]> is behind <c1,CAM_FRONT,[533, 251]>, <c6,CAM_BACK,[495, 796]> is behind <c1,CAM_FRONT,[533, 251]>, <c3,CAM_FRONT,[562, 273]> is in front of <c2,CAM_FRONT_LEFT,[159, 400]>, <c4,CAM_FRONT,[362, 330]> is in front of <c2,CAM_FRONT_LEFT,[159, 400]>, <c6,CAM_BACK,[495, 796]> is behind <c2,CAM_FRONT_LEFT,[159, 400]>, <c5,CAM_FRONT_RIGHT,[909, 393]> is on the back right of <c2,CAM_FRONT_LEFT,[159, 400]>, <c5,CAM_FRONT_RIGHT,[909, 393]> is behind <c3,CAM_FRONT,[562, 273]>, <c6,CAM_BACK,[495, 796]> is behind <c3,CAM_FRONT,[562, 273]>, <c4,CAM_FRONT,[362, 330]> is on the back left of <c3,CAM_FRONT,[562, 273]>, <c6,CAM_BACK,[495, 796]> is behind <c4,CAM_FRONT,[362, 330]>, <c5,CAM_FRONT_RIGHT,[909, 393]> is on the back right of <c4,CAM_FRONT,[362, 330]> and <c6,CAM_BACK,[495, 796]> is behind <c5,CAM_FRONT_RIGHT,[909, 393]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/82024d9f5f7f46dba436cd93c27173c7_8f800722c7364cf9880cacf850b8550b.jpg"
] | What is the relative positioning of the important objects in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
370381 | Only one of the boxes is a traffic sign. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/2bbae5e654224cbeb5884c471e2ad05e_0f2608005cc04b00b9573d903f794633.jpg"
] | Are <c4,CAM_FRONT,[508, 183]> and <c1,CAM_BACK,[503, 787]> traffic signs? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
148160 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/b0644660d7864451bfc4b3432bb5b47e_61f54f56a9824c0fb3dfc9785e6b110f.jpg"
] | Is <c1,CAM_BACK,[509, 836]> an object that the ego vehicle should consider in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
376139 | B | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/7e3a6bdd6c6f4c8fb018cff404974446_2c5ddecd3e2048598d9e8259e9765874.jpg"
] | What is the moving status of object <c3,CAM_FRONT,[350, 306]>? Please select the correct answer from the following options: A. Bending over. B. Turn right. C. Turn left. D. Reverse parking. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
34170 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/0c601ff2bf004fccafec366b08bf29e2_15ab3872150347678f4efca8c640ce75.jpg"
] | Is <c6,CAM_BACK_RIGHT,[753, 781]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
247866 | Keep going straight. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/2f093cad7209436aa06e103bfe062857_9393cc8d341740309424e20b8f34dcab.jpg"
] | What is the future state of <c3,CAM_BACK,[503, 815]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
352133 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/c04cdbbe744c4da08d3df61c47d5cc94_6fa7cf634e90492dadf183025f5effa5.jpg"
] | Would <c1,CAM_FRONT,[528, 303]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
72831 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/3b5f3391078e48ac92600cbdfa4ccbe6_f709cc87399044be9937b641caa15d30.jpg"
] | Would <c2,CAM_FRONT,[488, 281]> be in the moving direction of the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
60728 | <c2,CAM_BACK_LEFT,[117, 862]> is on the back left of <c1,CAM_FRONT_RIGHT,[914, 297]>. <c3,CAM_BACK,[493, 864]> is on the back left of <c1,CAM_FRONT_RIGHT,[914, 297]>, and <c3,CAM_BACK,[493, 864]> is on the back right of <c2,CAM_BACK_LEFT,[117, 862]>. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/9e7d5adad1514dd79292e9318b7a443c_1289e50eba7243a1866b13bb81d445f1.jpg"
] | What is the relative positioning of the important objects in the current scene? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
18970 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/8fe2dd6f533646b39c8e81316597c52b_c9390af3852e48139719a189a9d4df59.jpg"
] | Is <c2,CAM_FRONT,[502, 287]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
289818 | There are three cars behind the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/dbe2eceb3e4649289366f674c2ba4274_43ed7234ea704e7b8e6fd524775473b3.jpg"
] | What are objects to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
248730 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/cc8c0bf57f984915a77078b10eb33198_b73152cab88f49d9ba195da81fde1809.jpg"
] | Are there parked cars to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
346333 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/d4d2faa310a748dda508d15c6d5c4244_d233fc65c0ab402683ec447b567c3e66.jpg"
] | Will <c2,CAM_FRONT,[493, 292]> be in the moving direction of <c3,CAM_FRONT_RIGHT,[755, 329]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
286020 | Black SUV. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/08be42eb2186411d8e2201225329f1c6_f52a1f9194234e268de40205ffc00b65.jpg"
] | What is the visual description of <c2,CAM_BACK_LEFT,[100, 828]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
174749 | Back up. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/3ace9decdd1f4ea0915b9862c95ca128_707adf9db2084990973d30ee2f2a04db.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c2,CAM_BACK,[479, 869]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
45341 | D | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/2ba06ef7b3b44cbfad6d771981d86af9_b1d5f0b98bdf40cca19629e0f4643fda.jpg"
] | Predict the behavior of the ego vehicle. Please select the correct answer from the following options: A. The ego vehicle is steering to the left. The ego vehicle is driving very fast. B. The ego vehicle is slightly steering to the right. The ego vehicle is driving with normal speed. C. The ego vehicle is slightly steering to the left. The ego vehicle is driving very fast. D. The ego vehicle is going straight. The ego vehicle is driving with normal speed. Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
114774 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e15260c4ed5b4505900d497ef92ddd2d_321c11c104bf4a279869bb2ff16ee681.jpg"
] | Are there traffic cones to the back left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
227228 | There is one pedestrian and one car to the front left of the ego car. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/1e3f0bd8063a4084b7104345b20bfb80_8e82c5d9c1e340bea289068d0eb35163.jpg"
] | What are objects to the front left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
178217 | Low. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/e333874a12d64a7d956f23fe6a4c5c62_b79a659b05cf433ba7ad3c0cde27bce0.jpg"
] | What is the probability of colliding with <c2,CAM_BACK,[512, 794]> after the ego vehicle goes straight and keeps the same speed? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
26770 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/bfe368f8c6a44efcb323a82f6974a9a6_8937b75340024a919b5f9b4c86608bdd.jpg"
] | Are there moving construction vehicles to the front left of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
168579 | Braking suddenly, backing up. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/ceca737af78440118b7b394a54723454_97e92f1b31324e929e2759282f3d1b5a.jpg"
] | In this scenario, what are dangerous actions to take for the ego vehicle? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
243817 | None, no, none. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/63252988084b4943a50734990c47e039_e861625c90454c688480e4ae9c6b0bba.jpg"
] | Which object is most likely to be occluded by <c5,CAM_FRONT_RIGHT,[747, 319]>? Would this object affect the ego vehicle? Based on this object, what action of the ego vehicle is dangerous? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
195666 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/df0c3cb16cde410eba8eea191765515c_c7228a0beb9b4f86b33c0c5493ccac00.jpg"
] | Are there parked trucks to the front right of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
137738 | Accelerate and go straight. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/1e9283d1ce98428aaa3b851487eb4f58_c1316ee8828443cd8ae8a964a4c186d7.jpg"
] | What actions taken by the ego vehicle can lead to a collision with <c2,CAM_FRONT,[500, 274]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
3301 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/2422d2652f4f42298a389a6e0dfef307_6f6485a3d7bd4695bc7c7cc21651b4fe.jpg"
] | Are there barriers to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
62140 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/9056ac9944e846109b00e31cc96259cb_0bb96ca9999a4ac79831ab7156ea173a.jpg"
] | Is <c5,CAM_BACK_LEFT,[196, 799]> a traffic sign or a road barrier? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
176368 | No. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/ef5f945852e84fcc88cab506746a6636_0edb067ee1bf4441af862a39e2da2591.jpg"
] | Will <c2,CAM_BACK,[515, 866]> change its motion state based on <c1,CAM_FRONT_LEFT,[142, 308]>? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
177258 | Yes. | [
"/slurm/daizhi_fan/datasets/MLLM-CL/domain/AD/images/drivelm/stitch/019b370afb444a2e8ce85654898307c8_be7cbdbdf0bc47ecbe32cc758796bcd6.jpg"
] | Are there moving pedestrians to the back of the ego car? Objects are encoded using <c,CAM,[cx,cy]>, where c is the identifier, CAM indicates the camera where the object’s center point is situated, and x, y represent the horizontal and vertical coordinates of the center point of the 2D bounding box. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.